Secondary Logo

Journal Logo

Technology, Computing, and Simulation: Original Laboratory Research Report

Intranasal Medication Administration Using a Squeeze Bottle Atomizer Results in Overdosing if Deployed in Supine Patients

Goldhammer, Jordan E. MD*; Dobish, Mark A. MD*; McAnulty, Joshua T. MS*; Smaka, Todd J. MD; Epstein, Richard H. MD

Author Information
doi: 10.1213/ANE.0000000000001686
  • Free

Vasoconstrictors and local anesthetics are commonly applied to the nasal mucosa using squeeze bottle atomizersa to reduce edema, limit bleeding, and provide analgesia in preparation for nasal intubation, nasogastric tube placement, and endoscopic sinus surgery. Despite widespread use of these devices, there are few clinical guidelines that address technical details related to safe administration of intranasal medication.

Drugs administered intranasally are highly bioavailable, and absorption is rapid. In an animal model of nasal absorption, bioavailability of medications with a molecular weight of <1000 g/mol approached 100%, and peak blood levels were reached within 30 minutes.1,2 Consequently, excessive doses administered to the nasal mucosa have resulted in serious adverse outcomes. For example, intranasal phenylephrine hydrochloride (HCl; molecular weight, 162.2 g/mol) has been associated with mydriasis, hypertension, bradycardia, cardiovascular collapse, intracerebral hemorrhage, pulmonary edema, and death.3–8

The purpose of this study was to quantify, via mannequin simulation, the amount of liquid delivered to the nasal mucosa when the patient is in the supine position compared with the upright, or sitting, position. We hypothesized that the volume delivered per spray would be larger when the spray bottle was operated in the supine position compared with the upright position. We also sought to determine administration parameters for intranasal drug delivery using a dip tube nasal squeeze bottle that would reliably provide the desired amount of atomized liquid per spray, and thus avoid the potential for inadvertent excessive drug administration.

METHODS

This was an in vitro study involving simulated intranasal administration of liquid from manually operated dip tube with atomizer nasal squeeze bottles. Six 25-mL bottles (Wheaton Industries Inc, Millville, NJ) were acquired from our central inpatient pharmacy. For testing, each bottle was filled with 15 mL of sterile water, matching the volume of aqueous 3% lidocaine HCl 0.25% phenylephrine HCl solution dispensed by our pharmacy for clinical use with these devices. A convenience sample of 10 anesthesia residents with perioperative experience administering nasal medications to patients was studied. Preliminary testing revealed that each deployment of the manual squeeze bottle in the vertical position resulted in the atomization of approximately 50 μL of liquid. Because of the small amount of liquid deployed per spray, 40 sprays were used for each mannequin trial in the upright position. Forty sprays were necessary to obtain a volume that was measurable on a scale accurate to 0.1 g. In the supine position, the delivered volume was approximately 0.5 mL, so 4 sprays were used for the supine mannequin test. An AirSim Advance Bronchi (TruCorp, Belfast, Ireland) mannequin was placed in both the upright (90° elevation) and supine (0° elevation) position (Figure 1). Subjects were instructed to insert the tip of the device into the nostril of the mannequin and administer the sterile water solution by squeezing the bottle with the same force as used in clinical practice. Volume per spray was measured by subtracting the weight of the bottle after each trial from the weight before the trial, converting this difference into milliliters of water (1 mg/mL) and then dividing by the number of sprays. An Acculab 333 digital scale (Sartorius AG, Göttingen, Germany), accurate to 0.1 g, was used for these measurements. Subjects alternated between administration to the mannequin in the upright (40 sprays) and supine (4 sprays) position for a total of 3 replicates in each position. The phenylephrine dose that would have been administered per spray was calculated based on a 0.25% phenylephrine solution.

F1
Figure 1.:
Simulated administration method. A, Orientation of the mannequin when testing in the upright position. B, Orientation of the mannequin when testing in the supine position. When the mannequin was in the upright position, the angle of the squeeze bottle was approximately 45°; whereas, in the supine position, the angle was approximately −45°.

Nine spray bottles were filled with 15 mL of sterile water and tested in replicates of 3 at 0°, 15°, 30°, 45°, and 90° angles to determine the relationship between the angles of administration on volume per spray delivered by a single subject (Figure 2).

F2
Figure 2.:
Determination of effect of administration angle on volume delivered per spray. A ring clamp was attached to the upright pole with an adjustable aluminum platform. With the bottle aligned with the platform at angles of 0°, 15°, 30°, 45°, and 90° elevations, the volume of liquid displayed per spray was determined (see Methods for details of the measurement technique).

The effect of the initial volume in the squeeze bottle on the amount dispensed per spray at 90° was studied using 20, 15, 12.5, and 10 mL of sterile water. Nine trials at each fill volume were recorded, and the volumes delivered per spray were calculated as described above by a single subject.

Statistical Analysis

The mean volume per spray among the average of the 3 replicates for each provider was calculated in the supine and upright positions, and the 2-sided Student t test was used to compare the difference between the means of the 2 groups. Similarly, the mean volume per spray at 0°, 15°, 30°, and 45° was compared with 90°, and the mean volume per spray at 20, 15, and 12.5 mL was compared with 10 mL. Data were analyzed using Microsoft Excel 2011 (Microsoft, Redmond, Wash). P values involving multiple comparisons were adjusted using the method of Holm-Bonferroni, with the corrected P < .05 required for significance. A sample size of 7 subjects in the supine and the upright groups was required to provide 90% power at α = .05 for the 2-sided Student t test to detect a 0.5-mL difference with a pooled standard deviation of 0.25 mL.

RESULTS

The mannequin position significantly affected the mean volume delivered per spray (Table, Figure 3). In the supine position, the mean volume was 0.56 ± 0.22 mL; in the sitting position, the mean volume was 0.041 ± 0.02 mL (difference, 0.52 mL; 95% confidence interval [CI], 0.37–0.67 mL, P < .001). Converting the spray volume to the dose of phenylephrine in a 0.25% solution, the mean difference in dose per spray between supine and upright mannequin position was 1300 ± 525 mcg (95% CI, 925–1675 mcg, P < .001; Table).

T1
Table.:
Effect of Mannequin Position on Delivered Spray Volume
F3
Figure 3.:
Notched box plot (with data points) comparing volume per spray in the supine and upright mannequin positions. The median volume per spray in the supine position was 0.56 mL compared with 0.041 in the upright position. The wider interquartile range in the supine position, compared with the upright position, indicated the increased variability during supine testing. There was one far outlier in the supine group, likely because of a more vigorous application of force by this subject, compared with the others.

Administration with a delivery angle of 0°, 15°, and 30° resulted in a larger volume per spray than when the bottle was oriented at the reference 90° angle (95% CI, for differences in milliliters from the reference value = 0.59 to 0.88, 0.18 to 0.26, and 0.03 to 0.06, respectively; P < .001 for all comparisons; Figure 4). The mean volume dispensed per spray at 45° (0.032 ± 0.006 mL) was not different from the volume delivered at 90° (0.030 ± 0.005 mL; 95% CI for difference, −0.003 to 0.008 mL; P = .34; Figure 4).

F4
Figure 4.:
Effect of dispensing angle on mean volume ± standard error per spray. Bottles were aligned as in Figure 2, and the volume per spray was determined (see Methods for details). P values refer to the difference between the reference volumes per spray at 90°, adjusted for 4 comparisons using the Holm-Bonferroni method. Below 45°, volumes dispensed were significantly larger than those at 90°. P values displayed are the corrected values for <.05 and the uncorrected values for those ≥.05.
F5
Figure 5.:
Effect of fill volume in the squeeze bottle on mean volume ± standard error per spray, measured at a 90° angle. P values represent the comparison of the differences in the volume per spray among the 20-, 15-, and 12.5-mL fill volumes and the amount dispensed per spray with a 10-mL fill volume, adjusted for 3 comparisons using the Holm-Bonferroni method. P values displayed are the corrected values for <.05 and the uncorrected values for those ≥.05.

Measured at 90°, a larger volume was dispensed per spray when the squeeze bottle was filled with 20 mL or 15 mL when compared with filling the bottle with 10 mL (95% CI for differences −0.041 to −0.014 mL and −0.012 to −0.0032 mL; P = .001 and .002, respectively; Figure 5). The mean volume dispensed per spray when the bottle was filled with 12.5 mL was not different from the volume dispensed when the bottle was filled with 10 mL (95% CI for difference = −0.005 to 0.0033 mL; P = .73; Figure 5).

DISCUSSION

We determined that when a dip tube squeeze bottle atomizer is used to administer medication during simulation testing, a 14-fold increase in volume is delivered to the nasal mucosa when the spray is administered with a mannequin in the supine position compared with the upright position (difference = 0.52 mL; 95% CI, 0.37 to 0.67 mL, P < .001). This finding is highly relevant clinically because practice variability results in administration of nasal medication in 1 of 2 settings: (1) in the upright position before induction of anesthesia; or (2) in the supine position after anesthetic induction. In the supine setting, 1400 mcg of phenylephrine, a large overdose, will be delivered to the nasal mucosa per spray when a 0.25% solution is used. Considering that bilateral dosing and multiple sprays per nostril are often employed, this can result in a massive phenylephrine overdose.

A clinical advisory report from New York was published following multiple intraoperative complications associated with nasal phenylephrine.3 The Phenylephrine Advisory Committee suggested: (1) restricting the use of topical phenylephrine; (2) that all medication delivery be administered in a calibrated syringe; and (3) limiting the initial dose of phenylephrine to 0.5 mg in adults and 20 mcg/kg in children.9 Our investigation found that this recommended maximal dose was exceeded by nearly 300% per spray when a nasal spray bottle was used in the supine position. As a consequence of our study, we eliminated the distribution of 0.25% phenylephrine solution in dip tube atomizers to our operating rooms.

Our findings highlight the importance of the orientation of the bottle if such devices are used to deliver intranasal medication. The minimal angle to ensure that the intended dose is administered is 45°, but we strongly recommend administering the drug with the bottle at 90° because only a small change in angle will result in the administration of a stream of medication and a much larger dose (Figure 6). Placing the table in reverse Trendelenburg or raising the back of the bed are possible strategies that can be employed in patients who are supine, particularly those who are anesthetized, to achieve a safe angle of administration. In addition, care must be taken not to administer an excessive number of squeezes, especially when both nares are topicalized.

F6
Figure 6.:
Effect of administration angle on aerosolization of the liquid dispensed from a squeeze bottle. A, Medication dispensed with the bottle at 0° position. Note that the entire dose is delivered as a stream, with no aerosolization. B, Medication dispensed with the bottle at 90° position. Note that the entire dose is aerosolized as a fine mist.

If using 25-mL bottles similar to those tested, we recommend not exceeding 12.5 mL (ie, approximately 50% of the total volume of the bottle). For other size bottles, the manufacturer should be consulted regarding the recommended maximum fill volume for their particular devices. Alternatively, independent testing, such as what we have described, should be performed to determine this value.

There are several limitations to this study. First, our convenience sample consisted of resident anesthesiologists. We studied only residents who reported a history of nasal spray device use within a clinical setting; however, subconscious bias may have existed in how they performed the testing, given their subordinate position to the faculty investigators. In addition, because our subjects represent a convenience sample of trainees, results may not be quantitatively generalizable to all anesthesia provider groups. Second, we conducted our measurements in vitro, where the angle of administration could be easily measured and controlled. In practice, smaller angles than those resulting in complete atomization of the dose could be utilized, resulting in inadvertent overdosing. Third, there was considerable variability in the volumes dispensed with the mannequin in the supine position (Figure 3). From a safety perspective, it is the outlier behavior that is more relevant than the average performance, so our findings may, in fact, underestimate the potential danger of dip tube squeeze bottle atomizers. Fourth, somewhat smaller volumes per spray probably would have been administered in the upright position if we had done the initial testing with <15 mL, as was noted in the tests of the angle of administration on the volume dispensed. However, this would not have affected the volume delivered in the supine position because the entire dose was delivered as a stream. Thus, the relative increase in the volume delivered per spray in the supine position would have been even larger than what we found. Finally, mannequin testing was accomplished with sterile water to allow accurate conversion from the weight of dispensed solution to the volume of dispensed solution. In clinical practice, the density and viscosity of local anesthetic and vasoconstrictor solutions may affect the kinetics of spray bottle deployment. Nonetheless, although the absolute value of a solution dispensed may differ because of fluid mechanics, the relative difference in volume dispensed between supine and upright administration will remain consistent.

In conclusion, we found that an approximate 14-fold increase in volume will be delivered when a squeeze bottle atomizer is used with a mannequin in the supine position compared with the upright position. Given the reported toxicity from the use of intranasal medication and the repeated case reports of inadvertent overdosing when squeeze bottle atomizers are used in clinical practice, we recommend that these devices only be used with a patient in the sitting position and that care be taken not to administer excessive sprays. Ideally, all intranasal drugs should be administered with a precise, metered-dose device.

DISCLOSURES

Name: Jordan E. Goldhammer, MD.

Contribution: This author contributed to study design, conduct of the study, data collection, data analysis, and manuscript preparation.

Name: Mark A. Dobish, MD.

Contribution: This author contributed to study design, conduct of the study, data collection, data analysis, and manuscript preparation.

Name: Joshua T. McAnulty, MS.

Contribution: This author contributed to conduct of the study, data collection, and manuscript preparation.

Name: Todd J. Smaka, MD.

Contribution: This author contributed to study design, conduct of the study, data collection, and manuscript preparation.

Name: Richard H. Epstein, MD.

Contribution: This author contributed to study design, conduct of the study, data collection, data analysis, and manuscript preparation.

This manuscript was handled by: Maxime Cannesson, MD, PhD.

FOOTNOTES

aThe tested spray bottle is constructed as a dip tube with atomizer tip, similar in form and function to spray bottles available over the counter. The dip tube atomizer is fabricated to expel an air and liquid admixture, resulting in aerosolization of fluid. Air resides within an internal tube. When squeezed, liquid mixes with air inside the tube and is expelled as an aerosolized spray through the tip of the device. When the device is held in a position other than upright, air within the tube is displaced by liquid, therefore, resulting in decreased aerosolization.

REFERENCES

1. Dale O, Hjortkjaer R, Kharasch ED. Nasal administration of opioids for pain management in adults. Acta Anaesthesiol Scand. 2002;46:759–770.
2. McMartin C, Hutchinson LE, Hyde R, Peters GE. Analysis of structural requirements for the absorption of drugs and macromolecules from the nasal cavity. J Pharm Sci. 1987;76:535–540.
3. Cantu C, Arauz A, Murillo-Bonilla LM, López M, Barinagarrementeria F. Stroke associated with sympathomimetics contained in over-the-counter cough and cold drugs. Stroke. 2003;34:1667–1672.
4. Kalyanaraman M, Carpenter RL, McGlew MJ, Guertin SR. Cardiopulmonary compromise after use of topical and submucosal alpha-agonists: possible added complication by the use of beta-blocker therapy. Otolaryngol Head Neck Surg. 1997;117:56–61.
5. Greher M, Hartmann T, Winkler M, Zimpfer M, Crabnor CM. Hypertension and pulmonary edema associated with subconjunctival phenylephrine in a 2-month-old child during cataract extraction. Anesthesiology. 1998;88:1394–1396.
6. Singh PP, Dimich I, Shamsi A. Intraoperative pulmonary oedema in a young cocaine smoker. Can J Anaesth. 1994;41:961–964.
7. Bird DJ, Markey JR. Massive pulmonary edema in a habitual crack cocaine smoker not chemically positive for cocaine at the time of surgery. Anesth Analg. 1997;84:1157–1159.
8. Prielipp RC. Unilateral mydriasis after induction of anaesthesia. Can J Anaesth. 1994;41:140–143.
9. Groudine SB, Hollinger I, Jones J, DeBouno BA. New York State guidelines on the topical use of phenylephrine in the operating room. The Phenylephrine Advisory Committee. Anesthesiology. 2000;92:859–864.
Copyright © 2016 International Anesthesia Research Society