Otologic surgery in patients with HIV-1 and AIDS

ORIGINAL ARTICLES Otologic surgery in patients with HIV-1 and AIDS DARIUS KOHAN, MD, and RENATO J. GIACCHI, MD, New York, New York

Otologic disease in patients infected with HIV occurs frequently and usually represents rhinologic disease and associated eustachian tube dysfunction rather than manifestations of HIV infection. As in all patients, the decision to operate on an HIVinfected individual who would benefit from major otologic surgery is a balance between the risks of the procedure and the possible benefits to the patient. Many concerns regarding wound infection and healing have been raised. The objective of this study is to evaluate the outcome of otologic procedures in this population. The charts of 9 men and 4 women were reviewed. Seven patients (54%) met the Centers for Disease Control and Prevention criteria for AIDS. Patients with chronic otitis media (46%) underwent tympanomastoidectomies, and the cases of acute mastoiditis (31%) were managed with simple mastoidectomies. Other procedures included repair of cerebrospinal fluid leak (15%) and stapedectomy (8%). Two patients had early complications and died during their hospitalizations. Three patients had prolonged hospital courses requiring long-term antibiotics. These 5 patients underwent urgent procedures and were severely immunocompromised. Of the remaining 8 patients only 2 had AIDS, and all had an uncomplicated postoperative course. Six of these patients were followed up for more than 1 year, and only 2 developed subsequent otologic disease. (Otolaryngol Head Neck Surg 1999;121:355-60.)

It has been nearly 16 years since the first cases of AIDS were reported in Los Angeles among 5 young homosexual men. The latest statistics indicate that 641,086 From the Department of Otolaryngology, New York University School of Medicine. Presented at the Annual Meeting of the American Academy of Otolaryngology–Head and Neck Surgery, San Francisco, CA, September 7-10, 1997. Reprint requests: Darius Kohan, MD, Department of Otolaryngology, NYU School of Medicine, 550 First Ave, New York, NY, 10016. Copyright © 1999 by the American Academy of Otolaryngology– Head and Neck Surgery Foundation, Inc. 0194-5998/99/$8.00 + 0 23/1/95363

men, women, and children were diagnosed with AIDS through December 1997, and 390,692 have died of their disease.1 However, the number of people living with HIV infection is estimated at more than 1 million. The average time for acquired HIV infection until AIDS develops is approximately 10 years.2 Otolaryngologists are frequently called on to evaluate patients with HIV and AIDS. Nearly all patients will have head and neck manifestations during the course of their disease.3-5 Otologic manifestations, although less common, are still frequently encountered. Patients with HIV exhibit a wide variety of otologic disease. Often, the disease is not a true manifestation of AIDS but incidental otologic disease in HIV-infected patients.3-5 Eustachian tube dysfunction and obstruction caused by a nasopharyngeal mass may account for the otitis media with effusion that is frequently encountered in this population.6 In most cases conventional medical therapy results in resolution of symptoms. However, some HIV-infected patients may require surgical intervention, either urgently or electively as part of their disease. The objective of this study was to determine surgical outcome in the HIV-positive (HIV+) population undergoing otologic/neurotologic procedures excluding myringotomy and tube placement. METHODS AND MATERIAL A retrospective analysis of patient’s charts at New York University–Bellevue Hospital Medical Centers from January 1990 to December 1996 identified 13 patients who were known to be HIV+ at the time of surgery for otologic disease. Each patient underwent evaluation and treatment by members of the department of otolaryngology with appropriate consultation by infectious disease specialists. HIV status, diagnosis, indications for surgery, procedures performed, and outcome were all extracted from the hospital and office charts. Patients were classified as either being HIV+ or having AIDS, according to the most recent Centers for Disease Control and Prevention (CDC) criteria for diagnosis of AIDS.7 RESULTS

The charts of 9 men and 4 women were reviewed (Table 1). The average age was 38 years (range 32 to 50 years). Seven patients (54%) met the CDC criteria for 355

356

Otolaryngology– Head and Neck Surgery October 1999

KOHAN and GIACCHI

Table 1. Summary of study population Patient Age no. (y) Sex

HIV CD4 count status (106/L)

1

50

M

AIDS

5

2 3

37 34

M M

AIDS AIDS

6 0

4

36

F

AIDS

4

5

34

F

AIDS

110

6

36

F

HIV+

566

7

34

F

HIV+

640

8 9

32 34

M M

AIDS HIV+

>200 569

10

46

M

HIV+

250

11

32

M

AIDS

6

12

38

M

HIV+

>200

13

45 M 37.5

HIV+

300

MEAN

Diagnosis/surgical indication

AD acute pseudomonal mastoiditis AS acute fungal mastoiditis AD acute pseudomonal mastoiditis AS acute fungal mastoiditis

Procedure

Postoperative course

AD simple mastoidectomy

6 wk IV ceftazidime

AS simple mastoidectomy AD simple mastoidectomy

4 wk IV amphotericin 8 wk ciprofloxacin

AS simple mastoidectomy

Wound infection and postauricular dehiscence; facial paralysis at 6 wk Uneventful

AD COM with cholesteatoma

AD modified radical mastoidectomy AD COM with cholesteatoma AD modified radical mastoidectomy AS COM with cholesteatoma AS canal wall up tympanomastoidectomy AS TM perforation AS transcanal tympanoplasty AS TM perforation AS postauricular tympanoplasty AD TM perforation AD postauricular tympanoplasty AS swab trauma with CSF leak AS repair of CSF leak, middle ear obliteration Temporal bone fracture with AS repair of CSF leak, CSF leak middle ear obliteration Otosclerosis AS KTP laser stapedotomy

Uneventful Uneventful Uneventful Uneventful Uneventful Died 1 wk after surgery of massive hemoptysis and respiratory distress Uneventful Uneventful

AD, Right ear; AS, left ear; AU, both ears; IV, intravenous; COM, chronic otitis media; TM, tympanic membrane.

AIDS. The mean follow-up was 15.8 months (range 1 week to 58 months, median 6 months). The surgical indications were chronic otitis media (46%), acute mastoiditis (31%), traumatic cerebrospinal fluid (CSF) leak (15%), and otosclerosis (8%). The procedures performed included tympanomastoidectomy with cholesteatoma removal (23%), 2 of which were canal wall down; tympanoplasty (23%); simple mastoidectomy for drainage (31%); traumatic CSF leak repair with middle ear obliteration (15%); and stapedectomy (8%). There were no intraoperative complications. Two patients had early postoperative complications and died during the courses of their hospitalizations. Both patients were brought to the operating room urgently, carried a diagnosis of AIDS with CD4 counts less than 10 × 106/L, and had multiple opportunistic infections. Patient 11 had AIDS and was being treated for Pneumocystis carinii pneumonia on the medical service when he was referred for otorrhea after a cotton swab injury. He was diagnosed with a CSF leak. The CT scan revealed air in the cochlea, vestibule, and cranium (Fig 1). He was treated initially with bedrest, spinal drainage, and intravenous antibiotics but continued to have a leak. He was then brought to the operat-

ing room and was found to have a hypermobile stapes with CSF leakage from the anterior portion of the oval window. Because the injury resulted in a profound hearing loss, the oval window was patched, and the middle ear was obliterated. The spinal drain was removed on postoperative day 3. There was no evidence of CSF leakage, and the wound was healing well. However, on postoperative day 7 he died of massive hemoptysis and respiratory distress. The second mortality was in patient 4, who had acute fungal mastoiditis that did not respond to intravenous fluconazole and amphotericin B. A CT scan demonstrated an opacified left middle ear and mastoid with severe erosion of the scutum and ossicles (Fig 2). She underwent an extended mastoidectomy without complication. Cultures grew Aspergillus fumigatus. She tolerated the operation well, but postauricular dehiscence with drainage developed on postoperative day 30 and progressed to facial nerve paralysis 2 weeks later. At this point, she was reexplored, and the mastoid cavity was debrided of granulation tissue and fungal debris. The facial nerve was decompressed lateral to the geniculate, and she underwent a radical mastoidectomy. Once again after surgery, the wound did not heal and required debridement. The patient’s condition continued to dete-

Otolaryngology– Head and Neck Surgery Volume 121 Number 4

Follow-up

KOHAN and GIACCHI

Length of follow-up (mo)

Well healed; AU myringotomy and tubes at 3 mo for otitis media with effusion Well healed Well healed Reoperation of radical mastoidectomy and facial nerve decompression; wound did not heal and patient died Well healed

4.0 2.0 6.2 4.9 1.2

Graft perforation and recurrent drainage starting 17 mo after surgery Well healed

58.4

Well healed Well healed

22.5 26.1

Graft perforation and recurrent drainage starting 13 mo after surgery —

15.4

Well healed; no evidence of CSF leak

42.8

Complete closure of air-bone gap, normal bone line

14.7 15.8

6.9

0.2

riorate, and she died of P carinii pneumonia 5 months after the initial procedure. Three other patients underwent mastoidectomy for acute mastoiditis (patients 1 through 3). They were initially treated with prolonged intravenous antibiotic therapy and intensive aural care but failed to improve. All patients were severely immunocompromised with CD4 counts less than 10 × 106/L. CT scans of all patients demonstrated complete opacification of the mastoid air cell system and bony destruction, most frequently located at the posterior fossa plate. The operative findings in all these patients included granulation tissue and purulence involving the entire mastoid air cell tracts, as well as bony dehiscences with exposed dura. The organisms responsible for the infections were Pseudomonas aeruginosa in 2 patients and A fumigatus in the other. These 3 patients improved slowly and healed well with a prolonged course of appropriate postoperative antibiotic therapy. Patient 1 returned 3 months after surgery with persistent effusions and required bilateral myringotomy and tube placement. Three patients underwent tympanomastoidectomy for cholesteatoma (patients 5 through 7). Two of these patients had a previous history of otologic disease before HIV infection. One patient had 2 previous mas-

357

toidectomies, and the other gave a history of previous tympanostomy tube placement and tympanoplasty. All 3 of these patients had uncomplicated postoperative courses. Patient 5 met the criteria for AIDS with a CD4 count of 110 × 106/L but no history of opportunistic infections. She had an uneventful recovery and was lost to follow-up after 6 weeks. The other 2 patients did not have AIDS. Patient 6 was well until 17 months after surgery when she returned with a perforation of the graft and otorrhea. She was followed up for 40 months with intermittent otorrhea and refused further intervention. AIDS developed 3 years after the initial procedure. Patient 7 was asymptomatic at her visit 7 months after surgery. Three patients underwent tympanoplasty for chronic otitis media with tympanic membrane perforations, intermittent drainage, and conductive hearing losses (patients 8 through 10). One patient had AIDS, and all had uneventful postoperative courses. Patient 10 returned with recurrent otorrhea and perforation 13 months after surgery, at which time he had AIDS. Two patients in this series underwent repair of traumatic CSF leak. Patient 11 had AIDS and died 1 week after surgery. Patient 12 had a left temporal bone fracture after a fall. He had a complete sensorineural hearing loss and underwent a transmastoid repair with obliteration of the middle ear. The postoperative course was uneventful, and he has been followed up for 43 months without sequelae. The last patient underwent a laser stapedectomy for presumed otosclerosis. He did not have AIDS. He had no postoperative complications and was followed up for 15 months with complete closure of the air-bone gap. DISCUSSION

In a recent review, the HIV epidemic was estimated to affect approximately 0.3% of US residents.8 Among HIV+ patients immune status may be quantified by the CD4 lymphocyte count. HIV primarily infects and replicates within CD4–T-lymphocyte helper cells, thus destroying them. In vitro CD4–T-lymphocyte helper cells secrete numerous cytokines (including transforming growth factor-β and interleukin 2), promoting cellto-cell adhesion and deposition of fibronectin and leading to enhanced wound healing.9 Several medical reports attempted to correlate the severity of HIVinduced immunocompromise, as determined by lowered CD4 counts and diminished cytokine production, to postsurgical wound healing.10-15 We were unable to identify any otolaryngologic studies on this topic. Consten et al11,12 and Lord16 noted delayed wound healing after anorectal surgery in HIV+ patients with

358

KOHAN and GIACCHI

Otolaryngology– Head and Neck Surgery October 1999

Fig 1. Patient 11. Axial (A) and coronal (B) noncontrast temporal bone CT images revealing air in the vestibule (arrows). C, Axial noncontrast head CT with air in the anterior (asterisk) and middle (arrows) cranial fossae.

low CD4 counts. The same studies found that malnutrition and HIV-related wasting syndrome, generally known to retard wound healing, were not a significant predictor for poor wound healing if nutrition was quantified only by serum albumin levels. Semprini et al17 reported increased risk of surgically acquired infections and delayed wound healing after caesarean section in HIV-infected mothers with CD4 counts under 200 × 106/L. Safari et al18 published a conflicting report indi-

cating that the preoperative mean CD4 count was unrelated to wound healing after anorectal procedures. The study did, however, point out that the rate of wound healing was adversely affected if the HIV+ patients also had AIDS. Two recent studies on HIV+ trauma patients undergoing emergency surgery only hinted at the possibility of increased risk of severe postoperative wound infections but concluded that standard posttrauma surgical care and techniques suffice in promoting appro-

Otolaryngology– Head and Neck Surgery Volume 121 Number 4

priate and timely wound healing (Carrillo et al19 and Guth et al20). Otologic pathology in the HIV+ population is relatively common and responds appropriately to standard medical therapy. Surgical intervention is rarely required. In performing this study, we asked ourselves 2 questions: (1) Are the criteria for otologic surgical intervention the same regardless of HIV status? (2) Does HIV status determine the surgical outcome? Although our data are limited by the small patient population, which precludes statistical significance, they hint that surgical outcome correlates to severity of the HIV infection as measured by the CD4 count. The 6 HIV+ patients without AIDS (all with CD4 counts >200 × 106/L) healed without incident, and only 1 had a recurrence at 17 months. Among the 7 AIDS patients, 2 died, 3 had protracted and complicated postoperative courses, 1 had late disease recurrence at 13 months, and 1 (the only one in the group with CD4 counts >200 × 106/L) did well until being lost to follow-up after only 2 months. The patients with the most severe immunocompromise (patients 1 through 4), with CD4 counts less than 10 × 106/L, also had the sequelae of multiple opportunistic infections (such as cytomegalovirus colitis), and all were nutritionally depleted and somewhat wasted. Despite these obstacles, 3 of the 4 patients healed well and went on to recover after a prolonged course of antibiotics. There is a clear trend that patients with more advanced AIDS infection, with more immunocompromise and lower CD4 counts, have a worse surgical outcome. Patients who were HIV+, were healthy, and had adequate CD4 counts responded well to otologic surgery. With the advent of improved pharmacologic therapy, such as protease inhibitors, the HIV+ population is healthier. Many patients currently have adequate CD4 counts even if previously they had more severe immunocompromise.21,22 We believe that strict surgical criteria for otologic procedures should be applied to all patients, regardless of HIV status. Otolaryngologists must not withhold warranted surgical therapy because of the patient’s HIV status. The surgical outcome correlates with the degree of immunocompromise. Thus a more protracted and complicated postoperative course may be expected in patients with low CD4 counts and severe ear disease. These patients may require more intensive medical therapy in conjunction with our colleagues in infectious disease. To our knowledge this series of 13 HIV patients who had undergone otologic surgery is the largest published to date. In 1988 Kohan et al4 reported on 26 patients with AIDS and otologic disease. One patient in that

KOHAN and GIACCHI

359

Fig 2. Patient 4. Coronal noncontrast CT scan with soft tissue filling the middle ear and noticeable ossicular erosion (arrow).

series with cholesteatoma required a tympanomastoidectomy and had no reported complications. More studies are required to determine levels of immunocompromise, as measured by CD4 thresholds, above which elective otologic procedures may be performed without enhanced patient risk. If immunocompromise is severe, only mandatory surgery should be performed in combination with intensive pharmacologic medical management and enhanced nutrition. CONCLUSION

Otolaryngologists often participate in the care of patients with HIV infection and/or AIDS. Otologic disease is frequently encountered in this population, and occasionally surgical intervention is necessary. In comparing patients with HIV versus those with AIDS, we found that the more immunocompromised AIDS patients had more advanced otologic disease at presentation and had a more complicated postoperative course. All of the patients in this series who did not have AIDS, as well as most of those with AIDS, healed uneventfully. The decision to operate on an HIV+ individual, as in all patients, is a balance between the risks of the procedure and the benefits to the patient. Knowledge of the patient’s immune status should allow the surgeon to better inform the patient regarding the risks and benefits of surgery in relation to his or her underlying medical condition. This situation is analogous to the individualized counseling that occurs in patients with diabetes, cancer, cardiac, or pulmonary disease in whom the risks of surgery are potentially greater. Patients with HIV infection had a generally unremarkable surgical course. The potential for a more complicated course may be

360

Otolaryngology– Head and Neck Surgery October 1999

KOHAN and GIACCHI

estimated in relation to the patient’s immunosuppression as measured by the total CD4 count, history of opportunistic infections, and malnutrition.

11. 12.

REFERENCES 1. Centers for Disease Control and Prevention. HIV/AIDS Surveillance Report 1997;9(2):2-43. 2. Bender BS, Bender J. Surgical issues in the management of the HIV-infected patient. Surg Clin North Am 1993;73:373-88. 3. Lalwani A, Sooy CD. Otologic and neurotologic manifestations of acquired immunodeficiency syndrome. Otolaryngol Clin North Am 1992;25:1183-97. 4. Kohan D, Rothstein SG, Cohen NL. Otologic disease in patients with acquired immunodeficiency syndrome. Ann Otol Rhinol Laryngol 1988;97:636-40. 5. Morris MS, Prasad S. Otologic disease in the acquired immunodeficiency syndrome. Ear Nose Throat J 1990;69:451-3. 6. Stern JC, Lin P, Lucente FE. Benign nasopharyngeal masses and human immunodeficiency virus infection. Arch Otolaryngol Head Neck Surg 1990;116:206-8. 7. Centers for Disease Control and Prevention. 1993 Revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults. MMWR 1993;41(RR-17):1-19. 8. Karon JM, Rosenberg PS, McQuillan G, et al. Prevalence of HIV infection in the United States, 1984 to 1992. JAMA 1996;276:12631. 9. Wojciak B, Crossan JF. The effects of T cells and their products on in vitro healing of epitenon cell microwounds. Immunology 1994;83:93-8. 10. Diettrich NA, Cacioppo JC, Kaplan G, et al. A growing spectrum of surgical disease in patients with human immunodeficiency

13. 14. 15. 16. 17. 18. 19. 20. 21. 22.

virus/acquired immunodeficiency syndrome. Arch Surg 1991; 126:860-5. Consten EC, Slors FJ, Noten HJ, et al. Anorectal surgery in HIVinfected patients. Clinical outcome in relation to immune status. Dis Colon Rectum 1995;38:1169-75. Consten EC, Slors JF, Danner SA, et al. Severe complications of perianal sepsis in patients with human immunodeficiency virus. Br J Surg 1996;83:778-80. Wakeman R, Johnson CD, Wastell C. Surgical procedure in patients at risk of HIV infection—comments. J R Soc Med 1990; 83:315-8. Emparan C, Iturburu CM, Portugal V, et al. Infective complications after minor operations in patients infected with HIV: role of CD4 lymphocytes in prognosis. Eur J Surg 1995;161:721-3. Burke EC, Orloff SL, Freise CE, et al. Wound healing after anorectal surgery in human immunodeficiency virus–infected patients. Arch Surg 1991;126:1267-71. Lord RV. Anorectal surgery in patients with HIV: factors associated with delayed wound healing. Ann Surg 1997;226:92-9. Semprini AE, Castagna C, Ravizza M, et al. The incidence of complications after caesarean section in 156 HIV+ women. AIDS 1995;9:913-7. Safavi A, Gottesman L, Dailey TH. Anorectal surgery in the HIV+ patient: update. Dis Colon Rectum 1991;34:299-304. Carrillo EH, Carrillo LE, Byers PM, et al. Penetrating trauma and emergency surgery in patients with AIDS. Am J Surg 1995;170: 341-4. Guth AA, Hofstetter SR, Pachter HL. HIV and the trauma patient: factors influencing post operative infectious complications. J Trauma 1996;41:251-5 [discussions 255-6]. Boulton A. HIV trial stopped early after good results [letter]. BMJ 1997;314:699. McDonald CK, Kuritz DR. Human immunodeficiency virus type 1 protease inhibitors. Arch Intern Med 1997;157:951-9.

American Laryngological Association Officers and Councilors (1999-2000)

President: Gerald B. Healy, MD, Boston, MA Vice President/President-Elect: Harold C. Pillsbury III, MD, Chapel Hill, NC Secretary: Robert H. Ossoff, DMD, MD, Nashville, TN Treasurer: Robert T. Sataloff, MD, DMA, Philadelphia, PA Historian/Editor: Stanley M. Shapshay, MD, Boston, MA First Councilor: Robert W. Cantrell, MD, Charlottesville, VA Second Councilor: John A. Tucker, MD, Philadelphia, PA Third Councilor: Lauren D. Holinger, MD, Chicago, IL Councilor-at-Large: Gayle E. Woodson, MD, Memphis, TN Councilor-at-Large: Gerald S. Berke, MD, Los Angeles, CA