prevention of hyperbaric-associated middle ear barotrauma

BRIEF REPORT barotrauma hyperbaric oxygenation

Prevention of Hyperbaric-Associated

Middle Ear Barotrauma

From the Emergency Medicine Residency Program, Butterworth Hospital; and Michigan State University College of Human Medicine, Grand Rapids.

Received for publication March 4, 1992. Revision received July 6, 1992. Accepted for publication July 20, 1992.

Presented at the Society for Academic Emergency Medicine Annual Meeting in Toronto, Ontario, Canada; and the Fourth International Conference on Emergency Medicine in Washington, DC, May 1992.

Scott Carlson, MD Jeffrey Jones, MD, FACEP Michael Brown, MD, FACEP

Chris Hess, RN, MS

Study objective: To determine the efficacy of topical nasal decongestant in the prevention of middle ear barotrauma in patients undergoing hyperbaric oxygen therapy.

Design: Prospective, parallel, double-blind, randomized trial.

Set t ing : University-affiliated community hospital emergency department with hyperbaric oxygen facilities.

P a r t i e i p a n t s : Sixty patients undergoing hyperbaric oxygen therapy; 30 subjects in each treatment arm.

Intervent ions: After randomization, consenting patients were given two sprays of oxymetazoline hydrochloride or sterile water, 15 minutes before hyperbaric oxygen therapy. Collected data included patient demographics, ear examinations before and after hyperbaric oxygen treatment, and subjective ear complaints. The otoscopic appearance of the tympanic membrane was graded according to the amount of hemorrhage in the eardrum, with Teed scores ranging from 0 (symptoms only) to 5 (gross hemorrhage and rupture).

Resu]ts: The treatment groups were similar with regard to age, sex, and medical history. Ear discomfort during hyperbaric oxygen therapy was present in 63% (19 of 30) of those receiving oxymetazoline versus 67% (20 of 30) of the control group (P= .99). Likewise, both groups had similar Teed scores after hyperbaric oxygen therapy (P= .88). No adverse effects were noted.

Conclusion: The results of this pilot study suggest that topical decongestants may not be effective in preventing middle ear barotrauma during hyperbaric oxygen therapy.

[Carlson S, Jones J, Brown M, Hess C: Prevention of hyperbaric- associated middle ear barotrauma. Ann Emerg Med December 1992;21:1468-1471 .]

7 0 / 1 4 6 8 ANNALS OF EMERGENCY MEDICINE 21:12 DECEMBER1992

BAROTRAUMA Carlson et al

I N T R O D U C T I O N 0tic ba ro t r auma , commonly called middle ear squeeze or barotitis media, is the most prevalent medical problem associated with a i rplane travel , scuba diving, and hyperbar ic oxygen (HBO) therapy. 1 The in jury occurs when individuals are unable to equalize the pressure gradient between the middle ear and the external environment during ascent or descent. Symptoms include ear pain, t innitus, and t ransient conductive hearing loss. 1,2 Clinical signs of otic ba ro t rauma encompass tympanic membrane congestion or hemorrhage, hemotympannm, or membrane rupture . Resolution of symptoms may take up to four weeks, depending on the severity of the injury. 3

There are no reports in the h te ra ture of well-controlled chnical evaluations of sympathomimetic drugs for the prevention or t reatment of hyperbar ic-associa ted middle ear barotitis. Various sources, however, recommend prophylac- tic and therapeut ic use of oral or topical nasal decongestants to prevent and t reat middle ear ba ro t rauma. 4-6 Although supporting data are not cited, oxymetazoline (Afrin TM) and phenylephrine (Neo-Synephrine TM) frequently are specified as the topical agents of choice for the t reatment of this condition.4,7,8

Therefore, to evaluate the efficacy of topical oxymetazoline for the prophylaxis of hyperbar ic-associa ted middle ear ba ro t rauma, a randomized, double-bl ind clinical t r ia l was conducted.

M A T E R I A L S A N D M E T H O D S All adults requiring HBO therapy for acute carbon monoxide poisoning between October 1989 and F e b r u a r y 1992 were ehgible for par t ic ipat ion in this study. Subjects were identified through the emergency depar tment at But terworth Hospital , a large community teaching hospital with a mono- place hyperbar ic chamber.

Exclusion cri ter ia included subjects who were less than 18 years of age, had a known intolerance to oxymetazoline or other sympathomimetics, were pregnant or lactating, had any acute d isorder of the ears or upper respi ra tory t ract , or had evidence of neurologic dysfunction precluding them from making an informed decision. All conditions and proce- dures of the investigation were approved by the Human Rights Review Committee at our institution.

An otoscopic examination was performed on each pat ient by the emergency physician. After giving informed consent, subjects were given oxymetazoline hydrochlor ide (0.05%) nasal spray or sterile water following a double-bl ind proto- col. Treatment consisted of two sprays into each nostr i l at least 15 minutes before HBO therapy. Medications were packaged in identical plastic bottles that were identified only by a sequential number for coding purposes. The bottles were numbered randomly by computer. The identification key was kept by the pharmacis t and seen by investigators only at the terminat ion of the study.

A registered nurse or respi ra tory therapist t rained in HBO therapy controlled the hyperbar ic treatments externally. Oxygen was delivered at 2.4 atm absolute for 90 minutes.

After completion of the hyperbar ic t reatment , subjects were asked if they experienced any otologic symptoms, such as pain, pressure sensation, hearing loss, or tinnitus. Queries were also made as to side effects (ie, i r r i ta t ion of mucous membranes, sneezing). All subjects then underwent a second otoscopic examination by the same emergency physician. The appearance of the tympanic membranes was graded according to the Teed classification system. The Teed system classifies the otoscopic appearance of the tympanic membrane according to the amount of hemorrhage in the ea rdrum, with Teed scores ranging from 0 (normal) to 5 (gross hemorrhage and rupture) . 9 To minimize observer variability, a copy of the Teed classification system with photo- graphic examples was posted in the hyperbar ic chamber.

Subject characterist ics were analyzed (Z 2, Fisher 's exact test, t-test) to compare the mean age, sex, medical history, diving experience, and current use of decongestants. The Mann-Whitney U test was used to test the null hypothesis that the mean Teed scores of the two t reatment groups were equal. Confidence intervals based on the normal approxima- tion to the binomial dis tr ibut ion were calculated on the difference in the propor t ions of oxymetazohne and placebo subjects who had a Teed score of i or more and each of the four symptoms of middle ear baro t rauma. For all analyses, P < .05 was considered significant. Summary statistics and other analyses were performed using Statistical Analysis Systems (SAS Insti tute, Inc, Cary, North Carolina).

R E S U L T S During the study period, 61 patients satisfied the eligibility cri ter ia , consented to par t ic ipate , and were enrolled in the study. Of these, one pat ient was excluded from the final analysis because of poor documentation. Thir ty subjects were randomized to each of two t reatment groups. The two groups were similar with regard to mean age, sex, medical history, diving experience, and current use of decongestants (Table 1).

Table 1. Subject demographics

Oxymetazoline Placebo (N=30) (N=30) P*

Age (yr + SD) 44 + 17.4 44 _+ 16.7 1.0 Male (%) 18(60.0) 21 (70.0) .59 Smoker (%) 10 (33.3) 13 (43.3) .59 Previous ear problems (%) 4(13.3) 6(20.0) .70 Current use of decongestants (%) 2 (6.7) 1 (3.3) ,91 Experienced diver (%) 0 (0.0) 2 (6.7) ,44

*Based on Student's t-test and Fisher's exact test when appropriate.

DECEMBER1992 21:12 ANNALS OF EMERGENCY MEDICINE 1 4 6 9 / 7 1


The Teed score after HBO therapy was greater than zero in 46.7% (14 of 30) of the oxymetazoline group and 40.0% (12 of 30) of the placebo group. This difference was not found to be statistically significant using the Mann-Whitney U test (T = 903.5, P = .88). None of the subjects developed gross hemorrhage or rup ture of the tympanic membrane (ie, Teed score of 5).

Approximately the same propor t ion of patients in each t reatment group had symptoms of middle ear ba ro t r auma during HBO therapy (Table 2). Despite this ear discomfort, all the patients were able to complete the hyperbar ic dive. Three patients in the control group and two patients who received oxymetazoline required a slower dive descent because of ear pain. Statistical analysis of each pa ramete r in Table 2 showed no significant differences (P > .5).

The sample size of 60 cases was large enough to reject a 50% risk reduction of ba ro t r auma in patients t rea ted with oxymetazoline (a = 0.05, g = 0.20).

None of the study patients (experimental or control) repor ted any adverse reactions to the topical nasal spray.

D I S C U S S I O N The proposed effectiveness of sympathomimetic agents for the prevention or t reatment of middle ear ba ro t r auma is based on their effect on ~-adrenergic receptors in the respi- ra tory t ract mucosa. Adrenergic vasoconstriction results in shrinkage of swollen mucosal membranes and reduction of tissue hyperemia, edema, and congestion. These effects are well documented for the nasal passages and presumably occur in other areas of the respi ra tory t ract , including the eustachian tube.I°,11 The expected sympathomimetic action of topical oxymetazoline, therefore, should improve or maintain eustachian tube patency and allow equil ibrat ion of middle ear pressures. Accordingly, ear pain and discomfort associated with changing external pressures are prevented or decreased. Although supporting data have not been available, topical decongestants frequently are recommended for the prophylaxis of middle ear ba ro t r auma during hyperbar ic dives.7,12

In this study, both control and oxymetazoline-treated groups received identical hyperbar ic therapy for carbon monoxide poisoning. Oxymetazollne was not shown to significantly prevent middle ear ba ro t rauma (> 50% risk

Table 2. Treatment group results

Oxymetazoline Placebo Parameter (N=30) (%) (N=30) (%) P

Teed ___ 1 14(46.6) 12 (40.0) .80 Ear pain 9 (30.0) 6 (20.0) .55 Hearing loss 7 (23.3) 5 (16.6) .75 Blockage 16 (53.3) 13 (43.3) .61 Tin@us I (3.3) 2 (6.6) 1.0 Asymptomatic 11 (36.6) 10 (33.3) 1.0

reduction, B = 0.20). Some amount of ear discomfort (eg, pain, blockage, t innitus, hearing loss) was present in 67% of the control group versus 63% of those receiving oxymetazohne. The physical signs of ba ro t r auma were not~ with otoscopic examination. Using the Teed classification, we were unable to demonstrate a decreased incidence of tympanic membrane congestion or hemorrhage in the treat. meat group.

Our study had several potential limitations that must be taken into account in evaluating the data. The populat ion sample was nonconsecutive; only stable, cooperative adult patients without evidence of neurologic dysfunction were asked to par t ic ipate . The selection cri teria excluded those patients with acute disorders of the ears or respi ra tory trac who are more likely to respond to a decongestant. Observer var iabi l i ty regarding the Teed classification of tympanic membrane appearance was possible. Having sample photos of the different Teed classifications could only par t ia l ly alle- viate this potential bias. The sample size was small, and although we found no statistical difference, a type II e r ror is possible.

The middle ear is the most common site of ba ro t rauma , pr imar i ly because of the complexity of eustachian tube function. 6 The eustachian tube is normally closed and opens through the actions of the tensor and levator muscles of the palate. Under normal conditions, the egress of air from the middle ear is passive and automatic. 13 The reverse direction must have muscular action, such as swallowing or yawning, or insufflation by the Valsalva maneuver. As environmental pressure increases, the air in the middle ear and within the eustachian tube is compressed. When a certain pressure differential is reached (approximately 90 mm Hg), the cart i- laginous por t ion of the eustachian tube will f irmly collapse, and fur ther attempts at equalization are futile. This is why an experienced scuba diver does not "get behind" and autoinflates during descent. 13

The failure to equalize tympanic pressures during hyperbar ic therapy may be caused by poor function of the eustachian tube as a result of congenital, anatomic, or chronic pathologic conditions; acute pathologic changes in the eustachian tube, such as edema associated with upper respi ra tory infection; and incidental conditions, such as coma secondary to carbon monoxide poisoning. 2 Body posi- tion has been demonstrated to affect eustachian tube function under hyperbar ic conditions. The more horizontal the body is in space, the less efficient the tube. This observation is thought to result from venous congestion. 14

P rope r t reatment of middle ear ba ro t r auma begins with caution and prevention. Medical examinations before HBO therapy should emphasize nasal and eustachian tubal function. Par t i cu la r attention should be pa id to those individuals who have pre-existing signs of middle ear or nasal disease. These individuals, as well as those patients who are obtund- ed or unconscious, may warran t oral decongestants or pro- phylactic myringotomy.7

7 2 / 1 4 7 0 ANNALS OF EMERGENCY MEDICINE 21:12 DECEMBER1992


Other impor tant factors in the prevention of barot i t is media include slowing descent rates, abort ing the dive before tympanic rupture occurs, and avoiding forceful autoinflation at pressure. 7 Too often, ear clearing is taught as "pinch your nose and blow air into your ears ." This modified Valsalva maneuver not only is often ineffective but also can be danger- ous as a possible cause of round window rupture . Safer and more effective techniques are those that allow for gentle inflation of the middle ear without increases in in t ra thoracic pressure. Most patients can learn the Frenzel 's maneuver, which consists of pinching the nose, closing the mouth, and pushing the tongue against the soft palate to force air through the eustachian tubes into the middle ears. 12

Prophylact ic oral decongestants may be of benefit in certain situations to facilitate effective ear clearing. A recent randomized clinical t r ia l with scuba divers 6 demonstra ted that 60 mg oral pseudoephedr ine before open water diving decreased the incidence and severity of middle ear ba ro t r au - ma. The pseudoephedrine group had less ear discomfort (8% versus 32%) and smaller Teed scores after diving than the control subjects. Fu ture investigations might compare oral versus topical sympathomimetics for the prevention of barotitis media.

C O N C L U S I O N

The results of this pilot study suggest that topical oxymetazoline may not be effective in preventing middle ear discomfort and baro t rauma during HBO therapy. However, much larger sample sizes will be requi red to detect smaller (ie, 10% to 20%) reductions in baroti t is media by sympathomimetic agents. The routine practice of many hyperbar ic facilities of administering topical decongestants to prevent ba ro t r auma should be re-examined.

R E F E R E N C E S

1. Nakashima T, Itoh M, Sato M, et el: Auditory and vestibular disorders due to barotrauma. Ann Otol Rhinol Laryngo11988;97:146-152.

2. Schuknecht HF: Pathologyofthe Ear, ed 1. Cambridge, Harvard University Press, 1974, p 309-310.

3. Adams GL, Boies LR, Hilger PA: Fundamentals of Otolaryngology: A Textbook of Ear, Nose, and Throat Diseases, ed 6. Philadelphia, WB Saunders, 1990, p 94-95.

4. Schwartz GR: Principles and Practice of Emergency Medicine, ed 2. Philadelphia, WB Saunders, 1985, p 1590-1591.

5. Gross PL, Hobbs ET, Castronovo FP, et al: Environmental hazards, in Wilkins EW (ed): Emergency Medicine: Scientific Foundations and Current Practice, ed 3, Baltimore, Williams & Wilkins, 1983, p 180-181.

6. Brown M, Jones J, Krohmer J: Peeudoephedrine for the prevention of barotitis media: A controlled clinical trial in 120 scuba divers. Ann Emerg Med 1992;21:849-852.

7. Youngberg JT, Myers RAM: Complications from hyperbaric oxygen therapy? Ann Emerg Med 1990;19:1356.

8. Rodenberg H: Severe frontal sinus barotraurna in an airline passenger: A case report and review. J Emerg Med 1988;6:113-115.

9. Edmonds C, Freeman P, Thomas R, et ah OtologicalAspects of Diving, ed I. Sydney, Australian Medical Publishing Co, 1973.

10. Hoffman BB, Lefkowitz R J: Catecholamines and sympathomimetic drugs, in Gilman AG, Rail TW, Neis AS, et al (eds): Goodman and Gilman's The Pharmacological Basis of Therapeutics, ed 8. New York, Pergamon Press, 1990, p 215.

11. American Medical Association DeparLment of Drugs, Division of Drugs and Toxicology: Drug Evaluations Subscription. Chicago, AMA, 1991, vol I (Resp), p 1-29.

12. Jain KK (ed): Textbook of Hyperbaric Medicine, ed 1. Toronto, Hog refe & Huber, 1990, p 97.

13. Farmer JC: Ear and sinus problems in diving, in Bore AA, Davis JC (eds): Diving Medicine, ed 2. Philadelphia, WB Saunders, 1990, p 200-222,

14. Ornhagen HC, Tallberg P: Pressure equilibration ofthe middle ear daring ascent. Undersea Biomed Res 1981;8:219-227.

Address for reprints: Jeffrey Jones, MD, FACEP Department of Emergency Medicine 100 Michigan Avenue, NE Grand Rapids, Michigan 49503

DECEMBER1992 21:12 ANNALS OF EMERGENCY MEDICtNE 1 4 7 1 / 7 3

prevention of hyperbaric-associated middle ear barotrauma

Documents