Bone Marrow Transplantation (2002) 29, 257–261  2002 Nature Publishing Group All rights reserved 0268–3369/02 $25.00 www.nature.com/bmt

Infections post transplant Risk factors for post-stem cell transplant sinusitis AM Thompson1, M Couch2, ML Zahurak1, C Johnson1 and GB Vogelsang1 1

Department of Oncology, Johns Hopkins University School of Medicine, Baltimore, MD,USA; and 2Department of Otolaryngology, Johns Hopkins University School of Medicine, Baltimore, MD, USA

Summary: An understanding of the factors that place the posttransplant patient at increased risk for sinusitis would help identify patients likely to develop disease and possibly allow for interventions that would decrease the incidence or severity of sinus disease. This retrospective study investigates the ability of screening paranasal sinus computed tomographic scans (CTs), clinical history, and potential risk factors for sinusitis, including history of tobacco use, history of allergies or asthma, IgG level, history of sinusitis, remission status and acute graft-versus-host disease (GVHD) to predict post-transplant sinusitis. Medical records and sinus CTs of 100 allogeneic bone marrow recipients were reviewed. There was no increased risk of developing sinusitis post SCT for patients with significant disease on screening CT, symptoms at time of transplant, a history of tobacco use, asthma or allergies, low IgG level, history of sinusitis or for patients at high risk of relapse. Patients with GVHD were 4.3 times more likely than patients without GVHD to develop sinusitis post transplant (95% CI: 1.7–11.0, P = 0.002). Acute GVHD places patients at greater risk of developing sinus infections. Bone Marrow Transplantation (2002) 29, 257–261. DOI: 10.1038/sj/bmt/1703353 Keywords: sinusitis; stem cell transplant; GVHD

The immunosuppression associated with stem cell transplant (SCT) leaves transplant recipients susceptible to infections at many sites, including the sinuses. While the prevalence of chronic sinusitis in the adult population is thought to lie around 14%, the prevalence of sinusitis after transplant jumps to approximately 31%.1,2 In an effort to decrease such infections in SCT recipients, investigators have attempted to identify patients at increased risk of sinusitis. To date, a few retrospective studies have examined risk factors for sinusitis in the SCT population. In their 1995 study, Shubiya et al1 looked at the association between sinusitis post bone marrow transplant and variables such as Correspondence: Dr GB Vogelsang, Bunting-Blaustein Building, Room 2M-89, 1650 Orleans Street, Baltimore, MD 21231, USA Received 8 May 2001; accepted 11 October 2001

stem cell source, HLA compatibility, conditioning regimen, graft-versus-host disease (GVHD), disease status, and primary diagnosis. These investigators found an increased incidence of sinusitis associated with allogeneic compared to autologous transplantation. Billings et al3 studied the prognostic value of screening sinus computed tomographic scans (CTs) in pediatric stem cell transplant (SCT) recipients and found a correlation between sinusitis on screening sinus CT and post-transplant disease.3 In efforts to identify patients at risk for later disease and detect occult infection pre-SCT, some centers routinely obtain pre-SCT sinus scans. The cost–benefit relationship for such an approach has not been evaluated. No studies to date, however, have investigated the ability of sinus history to predict the development of post-transplant infection. Studies and undocumented observations have linked cigarette smoking, asthma, allergic rhinitis, selective IgG deficiencies, and sinus and nasal abnormalities to sinusitis in the non-transplant population.4–7 These factors that potentially predispose the pre-SCT recipient to sinusitis, also exist in the post-SCT recipient and hence may be helpful markers of patients likely to develop disease. A history of sinusitis or sinus symptoms identified during transplant work up may similarly signal patients with predisposing factors for sinusitis. This retrospective study investigates the ability of the sinus history and potential risk factors for sinusitis to predict post-transplant sinusitis. It also examines the ability of screening paranasal sinus CTs to identify patients at increased risk of post-transplant sinus infection.

Methods Subjects The study population was comprised of all patients of 18 years and older who underwent allogeneic bone marrow transplantation at Johns Hopkins Hospital over an 18 month period extending from 1 January 1988 to 30 June 1999. To be included in the study patients needed accessible screening sinus CTs and pre-SCT intake notes. One hundred and eleven allogeneic SCTs were performed over this 18 month period. Eleven patients were excluded because the screening sinus CT or intake note could not be located.

Risk factors for post-stem cell transplant sinusitis AM Thompson et al

258

Symptoms of sinusitis and risk factors for sinusitis The allogeneic SCT candidate evaluation, performed before SCT, was used to look for symptoms of sinusitis prior to bone marrow transplant and to assess for risk factors for sinusitis. The interval history of the patient after SCT was taken from the last note in the patient’s chart and was used to look for symptoms of sinusitis post SCT. Criteria for clinical acute sinusitis included complaints of nasal blockage or congestion, nasal discharge, hyposmia, facial pressure or pain, or terms that might indicate the presence of such symptoms. Patients with any of these symptoms were designated as having clinical sinusitis. Details of the information recorded on the presence or absence of sinus symptoms were noted. Forty-seven of the intake forms specifically remarked on the presence or absence of sinus symptoms. The presence or absence of underlying risk factors for sinusitis such as abnormal IgG levels, history of asthma, atopy or tobacco use was studied, as was history of sinusitis. IgG levels were not available on 10 patients and were not recorded for the 15 patients with multiple myeloma. Potential risk factors for sinusitis specific to bone marrow transplant patients including acute GVHD and remission status were also documented. Paranasal sinus CTs Information was obtained from pre- and post-bone marrow transplant sinus CTs. Pre-SCT sinus CTs were defined as those sinus CTs taken before the transplant as part of the work up for transplant and were done a median of 22 days prior to transplant. Post-SCT sinus CTs were performed as necessitated by the patient’s clinical picture and were available on 64 of the 100 patients. The degree of sinusitis was graded on CT scans according to two methods, one likely to achieve greater sensitivity and one likely to realize greater specificity. CTs were graded using a modified version of the method of Lund et al8 in which the left and right ethmoid, maxillary, frontal and sphenoid sinuses were each given a score from 0 to 2. A score of 0 denoted a clear sinus, 1 indicated partial opacity and 2 signified total or near total opacity, with opacity secondary to mucosal thickening or air fluid levels. The osteomeatal complexes were assigned a 0 or 2, indicating patency or occlusion, respectively. Based on this CT scoring system, patients were designated as having no (score of 0), minimal (score between 1 and 3), moderate (score between 4 and 10) or severe sinus disease (score of 11 to 20). The choice of cut off numbers dividing the scores into different categories of disease severity was arbitrary. The second grading system defined sinusitis as the presence of an air fluid level and/or total to near total opacity of a sinus. These scoring systems do not make a distinction between chronic and acute sinusitis. Statistical analysis Proportions were compared using the chi-square test or the Fisher’s exact test where appropriate. Logistic regression analysis was used to determine odds ratios for the development of radiologic evidence of sinusitis after transplant.9

Bone Marrow Transplantation

Computations were performed using the SAS (Statistical Analysis System) and all P values reported are two-sided.10 Results The characteristics of the study population including the age of the population and indications for transplant are shown in Table 1. The mean age was 44 years. Transplants were undertaken for a range of diseases, with non-Hodgkin’s lymphoma, acute myelogenous leukemia, acute lymphocytic leukemia, chronic myelogenous leukemia and multiple myeloma serving as the most common diagnoses. Preparative regimen was determined by diagnosis with patients transplanted for AML, MDS, multiple myeloma, and Hodgkin’s disease receiving Bu/Cy and lymphocyte-depleted grafts using elutriation, with 6 months of post-transplant cyclosporine. Three patients with aplastic anemia received unmanipulated grafts after cyclophosphamide and 1 year of cyclosporine. The remaining patients with hematological malignancies received Cy/TBI preparative regimens with lymphocyte-depleted grafts and cyclosporine. The risk of pre- or post-transplant sinusitis was not statistically significantly different by diagnosis or preparative regimen (P = 0.46). All patients engrafted promptly (median neutrophil recovery to ⬎500/␮l, 16 ± 3 days and platelet recovery 25 ± 7 days) and there were no engraftment failures. All patients had febrile neutropenia which was treated with empiric antibiotics. All patients were screened weekly for CMV to day 100 and there were no documented cases of CMV pneumonitis. The 39 patients found to be viremic were treated with induction and then maintenance gancyclovir to day 100, with growth factors given as needed to treat and prevent neutropenia. The frequency of radiographic evidence of sinusitis preand post-transplant as determined by the two grading systems is shown in Table 2. Twelve patients had radiographic evidence of moderate to severe sinusitis prior to transplant and 16 had air fluid levels and/or total to near total sinus opacification All patients with both symptoms and radiographic abnormalities received therapy for sinusitis pretransplant (n = 5). The majority of patients with radiographic abnormalities was treated prior to transplant (n = 15). Sixty-four patients had accessible post-SCT paranasal

Table 1

Baseline characteristics of the 100 study participants

Age (year)

Diagnosis Non-Hodgkin’s lymphoma Acute myelogenous leukemia Chronic myelogenous leukemia Multiple myeloma Acute lymphocytic leukemia Hodgkin’s disease Myelodysplastic syndrome Aplastic anemia Chronic lymphocytic leukemia Breast cancer

19–65 (mean: 44)

26 19 15 15 7 6 6 3 2 1

Risk factors for post-stem cell transplant sinusitis AM Thompson et al

Table 2 The frequency of radiographic evidence of sinusitis before and after transplant among the 100 study participants Paranasal sinus CT findings No CT obtained

Pre-SCT

Post SCT

0

36

Modified Lund et al criteria No sinusitis Minimals inusitis Moderate Severe sinusitis

51 37 12 0

11 24 24 5

Air fluid level and/or total or near total sinus opacification Present Absent

16 84

27 37

sinus CTs. Of these, 29 showed moderate to severe disease and 27 had air fluid levels and/or total to near total sinus opacification. Twenty-nine patients received therapy for post-transplant sinusitis. No patients developed infections suggestive of complications from untreated sinusitis, such as orbital cellulitis or brain abscess. A comparison of the two grading systems (modified version of the Lund et al grading system vs air fluid levels and/or total to near total sinus opacification), found in Table 3, shows considerable overlap between the group designated as having moderate to severe sinusitis and that with air fluid levels and/or total to near total sinus opacification (kappa pre-transplant = 0.59, post transplant = 0.75). In Table 4 the frequencies of the studied variables and their association with sinusitis on post-SCT CT are shown. Thirty-four patients had a history of tobacco use, 12 had a history of asthma or allergies, 14 had a history of sinusitis, and 11 had low IgG levels. No statistically significant correlation existed between history of tobacco use, sinusitis, allergy or asthma, or low IgG level and the findings of disease on post-SCT CT. The presence of symptoms prior to transplant, documented in 11 patients, and high risk relapse status, documented in 67 patients, also showed no statistically significant correlation with moderate to severe disease or the presence of air fluid levels or total to near total sinus opacification on post-transplant CT. Of the variables investigated only the presence of GVHD correlated with paranasal sinus disease after transplant. Forty-eight patients developed GVHD (stage I–IV). Patients Table 3 A comparison of two grading systems: grading sinus CTs using a modified version of the Lund et al system vs using the presence of air fluid levels and/or total to near total sinus opacification Radiographic findings

Frequency of air fluid level and/or total to near total sinus opacification

Screening sinus CT No to minimal disease Moderate to severe disease

8% (7/88) 75% (9/.12)

Post-transplant sinus CT No to minimal disease Moderate to severe disease

6% (4/35) 79% (23/29)

developing GVHD were 4.3 times more likely to develop clinically significant sinusitis as evidenced by moderate to severe disease on post-SCT paranasal sinus CT (95% CI: 1.7–11.0, P = 0.003). The association between sinus symptoms prior to transplant and the findings on screening sinus CT are shown in Table 5. Of the 12 patients with moderate to severe disease on screening sinus CT, five (42%) had sinus symptoms. Two (5%) of the 37 patients with minimal disease on screening CT and four (8%) of 51 patients with no disease on screening CT had sinus complaints. Twenty-five of 51 patients with no disease, 24 of 37 patients with minimal disease and four of 11 patients with moderate to severe disease had intake forms that did not specifically comment on sinus symptoms. Only the association between symptoms of sinusitis and moderate to severe disease on screening CT was statistically significant (P = 0.003).

259

Discussion This retrospective study attempted to identify risk factors for the development of post-transplant sinusitis. A review of the records and films of 100 adult SCT patients showed no statistically significant correlation between the findings on screening sinus CT and post-SCT paranasal sinus CT. In addition, no statistically significant correlation existed between low IgG levels, history of tobacco use, asthma or atopy, or remission status and the development of radiographic evidence of clinically significant sinusitis. Of the variables investigated, only GVHD had a statistically significant correlation with post-transplant sinusitis. The development of GVHD places patients at 4.3 times greater risk of having moderate to severe disease on CT post SCT (95% CI: 1.7–11.0, P = 0.003). Both the treatment of GVHD and the disease itself leaves patients immunosuppressed and more vulnerable to infection. Other factors found to place ICU patients at risk for sinusitis (intubations, nasogastric tubes etc) were associated with rapid mortality in the patients included in this study which likely explains the lack of association with sinusitis (data not shown).11 In contrast to the findings of this study, Shubiya et al found no association between post-transplant sinusitis and GVHD in their 1995 work.1 The conflict between the results in these two studies possibly stems from insufficient statistical power to detect a difference in the earlier work. This study reviewed 100 allogeneic SCT patients. Shubiya et al looked at 54 such patients. While Shubiya et al’s work documented no association between GVHD and sinusitis, a link between stem cell source and sinusitis was found. Shubiya et al documented an increased incidence of sinusitis in allogeneic vs autologous transplant recipients. Given the immune dysfunction and longer immune recovery time in allogeneic SCT, in which GVHD plays a prominent role, it is conceivable that GVHD was a confounder in the relationship between stem cell source and sinusitis found in Shubiya et al’s earlier study. The lack of a statistically significant correlation between screening sinus CT and post-SCT CT is also in contrast to the findings of past investigators. In their study of 51 pediatric patients, Billings et al3 found that screening sinus CTs Bone Marrow Transplantation

Risk factors for post-stem cell transplant sinusitis AM Thompson et al

260

Table 4

The frequency of the variables under study and their association with disease on post-SCT paranasal sinus CT

Characteristic

No. of patients with characteristic

Frequency of moderate to severe disease on post-SCT CT

P value

Frequency of air fluid levels and/or total to near total opacity on post-SCT CT

P value

34 64

32% (11) 27% (17)

0.55

29% (10) 25% (16)

0.64

12 88

17% (2) 31% (27)

0.50

8% (1) 30% (26)

0.17

History of tobacco usea Yes No History of allergy/asthma Yes No Low IgG levelsb Yes No Risk of relapse High Low

11 64

27% (3) 22% (14)

0.70

36% (4) 17% (11)

0.21

67 33

31% (21) 24% (8)

0.46

31% (21) 18% (6)

0.16

GVHD Yes No

48 52

44% (21) 15% (8)

0.002

42% (20) 13% (7)

0.002

Symptoms at time of intake Yes No

11 89

45% (5) 27% (24)

0.29

45% (5) 25% (22)

0.16

History of sinusitis Yes No

14 86

21% (3) 30% (26)

0.75

14% (2) 29% (25)

0.34

Findings on screening sinus CT No to minimal disease Moderate to severe disease

88 12

28% (25) 33% (4)

0.74

26% (23) 33% (4)

0.73

a

Smoking history was not available on two patients. Patients with multiple myeloma were excluded from the analysis. An additional 10 patients lacked information on IgG levels.

b

Table 5 Association between sinus symptoms prior to transplant and screening paranasal sinus CT findings Radiographic findings on screening sinus CT

Frequency of symptoms present

P value

Level of disease No disease to minimal disease Moderate to severe disease

7% (6/88) 42% (5/12)

0.003

Air fluid levels and/or total to near total opacity Absent Present

7% (6/84) 31% (5/16)

0.01

could identify patients at risk for later sinusitis. Shaw et al,10 studying risk factors for sinusitis after chemotherapy in 26 patients with hematological malignancies, found a link between abnormal screening sinus CTs and the development of disease after transplant. In both studies, the authors report a statistically significant difference between the development of post-transplant sinusitis in patients with no disease on screening sinus CT and those with any degree of sinus disease on screening sinus CT. However, no correlation between pre- and post-transplant CTs was found in our study (even when comparison was made between patients with no disease and patients with minimal to severe disease on sinus CT). The etiology of the conflicting findBone Marrow Transplantation

ings in the two studies is unclear. It possibly reflects a difference in predictive value of CTs in different populations: pediatric vs adult, patients prior to chemotherapy vs patients prior to SCT. The difference could also stem from the impact of the varying interventions that were performed in response to the initial CT findings. In this study, patients with significant disease on screening CT were generally treated prior to therapy. The extent to which interventions were made prior to treatment in the other two studies is unclear. Like any retrospective review, this study is limited by the thoroughness with which the initial data were collected and recorded. It is possible that patients with a history of asthma, atopy or sinusitis were missed because these health problems were not documented. Half the intake forms did not comment specifically on the presence or absence of sinus symptoms. Our data showed no statistically significant correlation between asthma and atopy, history of sinusitis and sinus symptoms at time of transplant and the severity of disease post transplant. However, any association between sinus disease and these history items could be muted by the inclusion of patients with these health problems into the group without such problems. Another limitation of this study lies in its broad definition of sinusitis. Clinical sinusitis was defined in this study as the presence of nasal blockage or congestion, nasal discharge, hyposmia, facial pressure or pain, or terms that

Risk factors for post-stem cell transplant sinusitis AM Thompson et al

might indicate the presence of such symptoms. However, these symptoms are also compatible with other diagnoses. In defining clinical sinusitis in such broad terms, this study may have overestimated the incidence of clinical sinusitis and hence potentially weakened the association between clinical sinusitis and other variables of interest. Likewise, no attempt was made to make a sharp distinction in this retrospective study between acute and chronic sinusitis, where different pathogens would be expected to play larger roles. The patients were felt to have acute sinusitis based on symptoms listed in the medical record in all but one case, where the patient was felt to have acute sinusitis in the setting of chronic sinus problems. Although the very nature of a retrospective study makes it likely that patients with sinusitis were missed, the lack of infections suggestive of untreated sinusitis and the cross-checking provided by using both radiographic and medical record data should ensure that the majority of patients was identified. Based on our study, approximately 12–16% of screening sinus CTs will demonstrate clinically significant disease. These findings are compatible with those of Shubiya et al1 which showed a prevalence of sinusitis of 10% in the pretransplant population. Such a frequency is also consistent with the reported nationwide prevalence of chronic sinusitis at 14%.2 While there are no clear guidelines as to optimal management of patients discovered to have sinusitis during pretransplant work up, it is generally held that sources of infection should be identified prior to immunosuppression. Screening sinus CTs will diagnose disease. However, the results of our study suggest that sinus disease could likely be uncovered in a more cost conscious manner. The statistically significant correlation between sinus symptoms and the presence of disease on screening CT suggests that sinusitis is not an occult infection in many patients and that screening for sinus symptoms is a way to determine which patients are more likely to have pretransplant disease. As a retrospective review of the data, this study was limited by the lack of information available on the presence of sinus symptoms in patients. The performance of a more thorough sinus history could potentially increase the sensitivity of

the history in identifying disease. Further investigation is needed to determine if detailed sinus histories can adequately identify pretransplant disease. A greater understanding of the implications of radiographically identified disease in a patient lacking symptoms is also needed to enable optimal management of the pretransplant patient.

261

References 1 Shubiya TY, Momin F, Abella E et al. Sinus disease in the bone marrow transplant population: incidence, risk factors, and complications. Otolaryngol Head Neck Surg 1995;113: 705–711. 2 Collins JG. Prevalence of selected chronic conditions: United States 1990–1992. National Center for Health Statistics 10 (194). 1997. 3 Billings KR, Lowe LH, Aquino VM, Biavati MJ. Screening sinus CT scans in pediatric bone marrow transplant patients. Int J Pediatr Otorhinolaryngol 2000; 52: 253–260. 4 Rachelefsky G, Goldberg M, Katz RM et al. Sinus disease in children with respiratory allergy. J Allergy Clin Immunol 1978; 61: 310–314. 5 Newman LJ, Platts-Mills TA, Phillips CD et al. Chronic sinusitis: relationship of computed tomographic findings to allergy, asthma, and eosinophilia. JAMA 1994; 271: 363–367. 6 Slavin RG. Nasal Polyps and sinusitis. In: Middleton E et al (eds). Allergy: Principles and Practice, 5th edn. Mosby: St Louis, 1998, pp 1027–1034. 7 Umetsu DT, Ambrosino DM, Quinti I et al. Recurrent sinopulmonary infection and impaired antibody response to bacterial capsular polysaccharide antigen in children with selective IgG-subclass deficiency. New Engl J Med 1985; 313: 1247– 1251. 8 Lund VJ, Kennedy DW. Quantification for staging sinusitis. Ann Otol Rhinol Laryngol 1995; S167: 17–21. 9 Cox DR. The Analysis of Binary Data. Methuen: London, 1970; SAS Institute Inc. SAS User’s Guide: Statistics, Version 5 edn. SAS Institute Inc: Cary, NC, 1985. 10 Shaw GY, Panje WR, Corey JP et al. Risk factors in the development of acute sinusitis in immunocompromised patients. Am J Rhinol 1991; 5: 103–108. 11 Weymuller EA, Bishop MJ, Santos PM. Problems associated with prolonged intubation in the geriatric patient. Otolaryngol Clin North Am 1990; 23: 1057–1074.

Bone Marrow Transplantation