Management of Non-Invasive Rhinosinusitis in the Immunosuppressed Patient Population

Transcription

1 The Laryngoscope VC 2015 The American Laryngological, Rhinological and Otological Society, Inc. Management of Non-Invasive Rhinosinusitis in the Immunosuppressed Patient Population Ashley M. Dao, MD; Shruthi K. Rereddy, BA; Sarah K. Wise, MD, MSCR; John M. DelGaudio, MD; Zara M. Patel, MD Objective: Rhinologists are seeing an increasing number of immunosuppressed patients. Currently, no treatment paradigm exists for treating acute and chronic noninvasive rhinosinusitis (ARS, CRS) in this growing population. This study aims to identify patient and treatment factors that affect rhinosinusitis outcomes in this vulnerable population. Study Design: Prognostic retrospective cohort study. Methods: Immunocompromised patients treated by rhinologists for ARS or CRS 10/2007 to 10/2012 were identified by rhinosinusitis diagnostic codes, codes for transplant, cancer, HIV, diabetes, and codes indicating immunosuppression in the intensive care setting. Associations between patient factors and outcome were analyzed by logistic regression. Associations between treatment and outcome were analyzed by Firth logistic regression. Results: A total of 132 subjects were identified. Of those, 90.9% had CRS and 9.1% had ARS; 12.9% were transplant patients; 47% were diabetic; 37.9% were cancer patients; and 16.7% were in the intensive care unit. Patients with higher American Society of Anesthesiologists (ASA) scores had decreased disease resolution (odds ratio [OR] 5 0.5, P ). Transplant patients (OR , P ), diabetics patients (OR 5 6.4, P ), cancer patients (OR 5 5.4, P ), and patients with prior medical therapy for rhinosinusitis (OR , P < 0.001) had increased disease resolution compared to immunosuppressed critical care patients. Patients treated with antibiotics alone had no statistically significant difference in disease resolution compared to those receiving no treatment. In contrast, treatment plans including surgery were associated with greater disease resolution. Conclusion: This data indicates that surgical treatment provides improved outcomes for patients presenting with acute exacerbations of rhinosinusitis related to their immunocompromised state. Given the limited study population, these findings may not apply to HIV-positive or ARS patients, and further study should be undertaken in these groups. Key Words: rhinosinusitis, chronic rhinosinusitis, acute rhinosinusitis, noninvasive rhinosinusitis, immunosuppressed, immunocompromised, diabetes, transplant, cancer, intensive care unit. Level of Evidence: 4. Laryngoscope, 125: , 2015 INTRODUCTION Physicians are seeing an increasing number of immunosuppressed patients due to advances in transplant medicine, increased longevity for HIV and cancer patients, and the 25 million adults in the United States living with diabetes. 1 Chronic rhinosinusitis (CRS) affects approximately 11.1 million Americans, and as of 2011, the cost associated with the treatment of CRS was estimated at $8.6 billion a year. 2,3 Currently, there are From the Emory University School of Medicine (S.K.R.), Atlanta, Georgia, U.S.A; and the Department of Otolaryngology Head and Neck Surgery, Emory University School of Medicine (A.M.D., S.K.W., J.M.DG., Z.M.P.), Atlanta, Georgia, U.S.A. Editor s Note: This Manuscript was accepted for publication March 5, Accepted for poster presentation at the Annual Meeting of the American Rhinologic Society, Orlando, Florida, U.S.A., September 20, The authors have no funding, financial relationships, or conflicts of interest to disclose. Send correspondence to Zara M. Patel, MD, Assistant Professor, Division of Rhinology Sinus and Skull Base Surgery, Department of Otolaryngology Head and Neck Surgery, 550 Peachtree St. NE, 11th Floor Atlanta, GA zara.m.patel@emory.edu DOI: /lary no established guidelines or recommendations regarding the treatment of rhinosinusitis in immunosuppressed patients and how it may differ from that of the general population; therefore, practice patterns vary widely across the country. 4 The only study suggesting a different approach to initial management in these patients is regarding balloon sinuplasty in patients with hematologic sequelae of their immunosuppression, thus diminishing their candidacy for traditional sinus surgery. 5 Recently, a study was published examining postsurgical outcomes in diabetics undergoing sinus surgery and found that they appear to be susceptible to different bacterial species and have poorer short-term quality of life. 6 There is evidence that suggests rhinosinusitis may be different in immunocompromised patients, and that along with the usual viral and bacterial agents that cause sinusitis in the general population there are additional and more virulent causes to be suspected and treated The purpose of this study is to examine patient factors, management decisions, and treatments received in the immunosuppressed patient population at one tertiary academic center and to analyze whether differences in management change patient outcomes. Although we set out to include HIV patients in this 1767

2 study, no HIV patients were included in this study due to algorithm artifact. MATERIALS AND METHODS After gaining approval from the Emory Institutional Review Board, patients treated by Emory rhinologists for acute rhinosinusitis (ARS) or CRS from October of 2007 to October of 2012 were initially identified using International Classification of Diseases (ICD-9) diagnostic codes. Subsequently, the patient sample was narrowed by filtering the initial group with ICD-9 codes for transplant, cancer, HIV, diabetes, and codes indicating immunosuppression in the intensive care setting (i.e., septic shock, inflammation-immunosuppression catabolism syndrome). Using this initial code screening, a total of 220 patients were gathered, and after careful chart review 132 were found to be truly immunosuppressed and had either ARS or CRS. Of the 88 patients excluded, reasons included cancer diagnoses that did not lead to immunosuppression, such as small skin cancers, or a coexisting invasive fungal sinusitis diagnosis. Patient characteristics including demographic factors, comorbidities, disease type (ARS or CRS), cause of immunosuppression, inpatient or outpatient setting, and prior management were recorded. Treatments including oral antibiotics, intravenous antibiotics, surgery, or combinations of those options were recorded, and subsequent outcomes were then also recorded for all patients. The primary outcome was disease resolution confirmed by computed tomography (CT) or nasal endoscopy. The secondary outcome was a subjective decrease in symptoms. Descriptive statistics were calculated for patient factors, treatment modality, and outcomes. Associations between patient factors and disease resolution were analyzed by logistic regression with P values obtained from Wald v 2 testing. Associations between treatment and disease resolution were analyzed by Firth logistic regression due to quasi-complete separation of data points. P values for these associations were also obtained from Wald v 2 testing. All analyses were performed using SAS version 9.3 (SAS Institute, Inc., Cary, NC). Significance level was set at P < RESULTS The final study sample consisted of 132 patients (76 men; 56 women) with a mean age of 52.9 years. A total of 90.9% of these patients had CRS, whereas only 9.1% had ARS. A total of 12.9% were transplant patients, 47% were diabetic, 37.9% had immunosuppression due to cancer therapy, and 16.7% had immunosuppression related to acute decompensation in the intensive care unit (Table I). Unfortunately, as noted above, computer algorithm artifact prevented us from collecting any data on HIV patients. Over 95% of the patients had a CT scan, over 70% of patients received oral antibiotics, and over 70% received either oral or IV antibiotics within 1 week of presentation (Table II). Average follow-up time was 7 months. Transplant patients (odds ratio [OR] , confidence interval [CI] 3.5, 143.8; P ), diabetics (OR 5 6.4, CI 1.4, 31.5; P ), and cancer patients (OR 5 5.4, CI 1.1, 29.6; P ) had improved disease resolution compared with critical care patients. In particular, the mean odds of resolution of sinusitis for transplant patients was 24 times that of nontransplant patients in the study population. Patients with prior medical therapy (prior episode) for rhinosinusitis (OR , CI 2.15, 15.9; P < 0.001) also had Variable TABLE I. Population Characteristics. Frequency (%) (for age, mean [SD]) Age 52.9 (18.1) Sex Male 76 (57.6) Female 56 (42.4) White 100 (75.8) Black 29 (22.0) Asian 1 (0.8) Hispanic 0 (0) Other/unknown 2 (1.5) Disease ARS 12 (9.1) CRS 120 (90.9) Setting Inpatient 28 (21.2) Outpatient 104 (78.8) ASA Score 1 6 (4.6) 2 38 (28.8) 3 61 (46.2) 4 26 (19.7) 5 1 (0.8) Comorbidities 2.9 (1.5) Immunocompromised State HIV/AIDS 0 (0) Transplant 17 (12.9) DM 62 (47.0) Cancer 50 (37.9) Critical care 22 (16.7) Prior medical treatment 56 (42.4) Prior surgical treatment 32 (24.2) ASA 5 American Society of Anesthesiologists; ARS 5 acute rhinosinusitis; CRS 5 chronic rhinosinusitis; DM 5 diabetes mellitus; SD 5 standard deviation. improved disease resolution versus those without prior medical therapy. Age, sex, race, and whether the patient was evaluated in the outpatient or inpatient setting did not positively or negatively influence disease resolution. Patients with higher American Society of Anesthesiologists (ASA) scores had decreased resolution of disease (OR 5 0.5, CI 0.28, 0.90; P ). Specifically, for every 1-point increase in ASA score, the mean odds of resolution of rhinosinusitis decreased by roughly 50% (Tables III and IV). Due to lack of sufficient data for subgroup analysis of oral versus intravenous antibiotics, patients who received either route of antibiotic therapy were grouped together for analysis of associations with outcome, although the proportion of patients on each route of antibiotic therapy is presented in Table II. Surgical treatment was strongly associated with disease resolution. For all sinusitis patients, P value for surgery alone was (OR , CI 3.8, > 1000) and surgery plus antibiotics was (OR , CI 1.01, 415.3). For 1768

3 TABLE II. Description of Treatment and Outcomes. Treatment Frequency (%) CT obtained or available 126 (95.5) Antibiotics (either alone, preoperatively or postoperatively)) Oral antibiotics 96 (72.7) IV antibiotics 38 (28.8) Antibiotics within 1 week 95 (72.0) Overall treatment None 10 (7.6) Antibiotics only 27 (20.5) Antibiotics and surgery 77 (58.3) Surgery only 18 (13.6) Outcome Symptomatic resolution 63 (47.7) Disease resolution confirmed 57 (43.2) by CT or endoscopy Continued disease 23 (17.4) Multiple treatments 23 (17.4) Lost to follow-up 23 (17.4) CT 5computed tomography. CRS patients, surgery alone had a P value of (OR , CI 2.5, > 1000). Interestingly, patients treated with antibiotics alone had no statistically significant difference in disease resolution compared to those receiving no treatment (Tables V and VI). These results are summarized graphically in Figures 1 and 2. TABLE III. Predictors of Resolution of Sinusitis (ARS and CRS). Variable OR (95% CI) P Value Age 1.00 (0.98, 1.03) Sex (female vs. male) 0.60 (0.26, 1.40) CRS vs. ARS 0.86 (0.16, 4.55) Inpatient vs. outpatient 0.60 (0.16, 2.23) ASA 0.50 (0.28, 0.90) Transplant 22.5 (3.52, 143.8) DM 6.42 (1.39, 31.5) Cancer 5.42 (1.07, 29.6) Critical care 3.30 (0.67, 16.3) Black vs. white 0.94 (0.34, 2.59) Asian vs. white Unknown/other vs. white 0.78 (0.03, 18.8) Prior surgical treatment 0.36 (0.12, 1.10) Prior medical treatment 5.84 (2.15, 15.9) < ASA 5 American Society of Anesthesiologists; ARS 5 acute rhinosinusitis; CI 5 confidence interval; CRS 5 chronic rhinosinusitis; DM 5 diabetes mellitus; OR 5 odds ratio. TABLE IV. Predictors of Resolution of Sinusitis (CRS Alone). Variable OR (95% CI) P Value Age 1.00 (0.97, 1.03) Sex (female vs. male) 0.49 (0.19, 1.26) Inpatient vs. outpatient 0.25 (0.05, 1.23) ASA 0.57 (0.30, 1.09) Transplant 98.9 (8.16, >1000) < DM 13.6 (1.93, 96.2) Cancer 12.3 (1.61, 93.7) Critical care 7.86 (1.05, 58.8) Black vs. white 1.25 (0.41, 3.82) Asian vs. white* Unknown/other vs. white 0.89 (0.04, 20.0) Prior surgical treatment 0.40 (0.12, 1.34) Prior medical treatment 6.44 (2.14, 19.4) < *No OR given for Asian vs. white comparison due to limited data (only 1 Asian subject). ASA 5 American Society of Anesthesiologists; CI 5 confidence interval; CRS 5 chronic rhinosinusitis; DM 5 diabetes mellitus; OR 5 odds ratio. DISCUSSION Immunocompromised patients with acute or chronic noninvasive rhinosinusitis present a difficult management challenge for the otolaryngologist because they are known to have possible different pathogenic etiologies and no dedicated treatment paradigm currently exists for their management. In 1999, Porter et al. reported that 66% of HIV patients surveyed had experienced at least one episode of sinusitis within the previous 6 months, and in 2001, Tarp et al. published on the presence and possible reactivation of Herpes virus type 1 to 8 in sinus aspirates of HIV-infected individuals. 11,12 Twenty years ago, first O Donnell et al. and then Fried et al. noted Pseudomonas to be an organism more likely found in transplant and other immunocompromised patients, and in 1996 cytomegalovirus was implicated in sinusitis in HIV patients by Marks et al. 7 9 A Brazilian group confirmed the importance of Pseudomonas as an opportunistic pathogen in transplant patients. 13 Milgrim et al. found Listeria monocytogenes and Candida albicans to be pathogenic species in HIV patients. 10 Even with this information available for over a decade regarding the prevalence and distinct additional pathogens involved in these immunocompromised patients, there has never been clinical research dedicated to the standardization and optimization of rhinosinusitis management for this patient population. Because of this dearth TABLE V. Effect of Treatment Modality on Resolution of Sinusitis (ARS and CRS). Treatment OR (95% CI) P Value Antibiotics only 4.02 (0.17, 93.8) Antibiotics and surgery 20.5 (1.01, 415.3) Surgery only 93.1 (3.78, >1000) ARS 5 acute rhinosinusitis; CI 5 confidence interval; CRS 5 chronic rhinosinusitis; OR 5 odds ratio. 1769

4 TABLE VI. Effect of Treatment Modality on Resolution of Sinusitis (CRS alone). Treatment OR (95% CI) P Value Antibiotics only 1.10 (0.03, 36.3) Antibiotics and surgery 14.2 (0.64, 314.0) Surgery only 66.4 (2.49, > 1000) CI 5 confidence interval; CRS 5 chronic rhinosinusitis; OR 5 odds ratio. of research, rhinologists across the country vary significantly in their selected treatment for these patients, with some preferring aggressive surgical management and others taking a more conservative approach. 4 Development of a standardized treatment protocol could lead to decreased treatment costs and overall improved patient outcomes. Protocols are generally most helpful when used as guidelines for treatment, while leaving room for the flexibility of clinical decision making based on individual case specifics assessed by treating physicians. Certain findings in our study were expected, such as prior medical therapy improving outcomes, as well as higher ASA scores as indicators of poorer prognosis. Increased ASA indicates a greater number or higher severity of comorbidities, which can negatively impact overall health and immune function. An interesting result was improved sinus disease resolution in diabetics, cancer patients, and transplant recipients compared to our immunocompromised intensive care unit group. There are several possibilities why this may be. Firstly, when patients in a chronically immunocompromised state present with sinonasal symptoms, there is widespread recognition of a possibly fatal outcome if those symptoms represent invasive fungal sinusitis. Thus, early recognition and lower threshold for aggressive management, even after the sinus disease is proven Fig. 1. Disease resolution by treatment modality. [Color figure can be viewed in the online issue, which is available at Fig. 2. Etiology: 12.9% of patients were transplant patients, 47% had diabetes mellitus, 37.9% were undergoing cancer treatment, and 16.7% were critical care patients. [Color figure can be viewed in the online issue, which is available at to be bacterial and noninvasive, may be playing a role. It is also true that when patients are in the intensive care unit, it is often for organ failure not classically associated with immunosuppression. It is only recently that we are starting to see more evidence about the effect on the immune system from a critically ill state, and physicians caring for these individuals may not prioritize the sinus disease in the face of other overwhelming medical issues. Persistent inflammation-immunosuppression catabolism syndrome is a term that has only been coined within the last 2 years, but in the intensive care literature has widely supplanted what used to be known as multisystem organ failure. 14 Also, when patients in the intensive care unit develop immunosuppression, it is often from acute decompensation. Thus, these patients may be even less capable of fending off infection than the chronically immunosuppressed. Lastly, patients in the intensive care unit are more likely to be ventilator dependent than the other patient groups in our study. Sinus symptoms will not be self-reported in ventilated, sedated patients, and are more likely to be incidentally found on imaging, or found on fever workups, leading to possible delays in diagnosis and management. The finding of improved sinus disease resolution with surgical treatment is certainly our most significant finding. It is important to note that we are using data from a tertiary care academic center, and that many of our patients were presenting in acute exacerbations of chronic sinus disease, which is the patient population that we are most concerned with learning more about. Obviously, in patients with CRS who are not experiencing acute exacerbations, these conclusions may not hold true. This association is likely due to the decreased ability of the immunocompromised patient to clear their disease using a heightened immunologic response with medical intervention alone as compared to the immunocompetent patient. Although uncomplicated noninvasive rhinosinusitis is not an emergent condition, we could perhaps take a lesson from our experience with invasive fungal disease and recognize the physiologic limitations that our immunocompromised patients have in mounting responses to opportunistic pathogens. This is further confirmed by our data showing that patients managed medically with antibiotics alone had no difference in disease resolution compared with those receiving no intervention at all. Therefore, earlier intervention with a

5 lower threshold for surgery may be indicated in these patients. We believe further data, collected in a prospective manner, will help clarify this point. Interestingly, although all rhinologists included in the study have treated HIV patients for rhinosinusitis over the study period, none of these patients were captured in the data collection process. This likely is due to artifact in the computerized data collection process. Although diagnostic codes were drawn as broadly as possible at the outset to capture all possible data points, some certainly were lost, which we recognize as a limitation of this type of retrospective study. CONCLUSION Our findings demonstrate that surgical treatment resulted in improved outcomes for our patient population of immunocompromised patients with noninvasive rhinosinusitis. Prior medical management was found to be a significant predictor of sinusitis resolution, whereas increasing ASA score and critical care status both were negative prognostic factors. Given our limited study population, with no HIV patients identified for analysis and less than 10% of our population with ARS, these findings may not apply to these patients, and further study should be undertaken in these subgroups. Acknowledgment Timothy Ryan, MD worked on the IRB protocol submission and compiled a portion of the appropriate ICD-9 codes for use in this study. BIBLIOGRAPHY 1. Centers for Disease Control and Prevention. National Diabetes Fact Sheet: National Estimates and General Information on Diabetes and Prediabetes in the United States, Atlanta, GA: US Department of Health and Human Services, Centers for Disease Control and Prevention; Bhattacharyya N. Functional limitations and workdays lost associated with chronic rhinosinusitis and allergic rhinitis. Am J Rhinol Allergy 2012;26: Bhattacharyya, Neil. Incremental health care utilization and expenditures for chronic rhinosinusitis in the United States. Ann Otol Rhinol Laryngol 2011;120: Patel ZM. Practice patterns regarding noninvasive rhinosinusitis in the immunosuppressed patient population. Allergy Rhinol (Providence) 2013;4:e151 e Wittkopf ML, Becker SS, Duncavage JA, Russell PT. Balloon sinuplasty for the surgical management of immunocompromised and critically ill patients with acute rhinosinusitis. Otolaryngol Head Neck Surg 2009; 140: Zhang Z, Adappa ND, Lautenbach E, et al. The effect of diabetes mellitus on chronic rhinosinusitis and sinus surgery outcome. Int Forum Allergy Rhinol 2014;4: O Donnell JG, Sorbello AF, Condoluci DV, Barnish MJ. Sinusitis due to pseudomonas aeruginosa in patients with human immunodeficiency virus infection. Clin Infect Dis 1993;16: Fried MP, Kelly JH, Strome M. Pseudomonas rhinosinusitis. Laryngoscope 1984:94: Marks SC, Upadhyay S, Crane L. Cytomegalovirus sinusitis. A new manifestation of AIDS. Arch Otolaryngol Head Neck Surg 1996;122: Milgrim LM, Rubin JS, Rosenstreich DL, Small CB. Sinusitis in human immunodeficiency virus infection: typical and atypical organisms. J Otolaryngol 1994;23: Tarp B, Kelsen J, Nielsen LP, et al. Herpesvirus type 1 8 in sinus aspirates from HIV-infected patients and immunocompetent individuals. Rhinology 2001;39: Porter JP, Patel AA, Dewey CM, Stewart MG. Prevalence of sinonasal symptoms in patients with HIV infection. Am J Rhinol 1999;13: Ortiz E, Ng RTY, Alliegro FC, Teixeira C, Muranaka EB, Sakano E. Microbiology of rhinosinusitis in immunosuppressed patients from the University Hospital. Braz J Otorhinolaryngol 2011;77: Gentile LF, Cuenca AG, Efron PA, et al. Persistent inflammation and immunosuppression: a common syndrome and new horizon for surgical intensive care. J Trauma Acute Care Surg 2012;72: