J Int Adv Otol 2019; 15(1): 18-21 • DOI: 10.5152/iao.2018.
5394
Original Article
The Association between Iron Deficiency and Otitis
Media with Effusion
Fatih Alper Akcan , Yusuf Dündar , Hümeyra Bayram Akcan , Derya Cebeci ,
Mehmet Ali Sungur , İlhan Ünlü
Department of Otorhinolaryngology, Düzce University School of Medicine, Düzce, Turkey (FAA, DC, İÜ)
Department Of Otolaryngology – Head & Neck Surgery, Wayne State University School of Medicine, Michigan, USA (YD)
Clinic of Internal Medicine, Ministry of Health Atatürk Hospital, Düzce, Turkey (HBA)
Department of Biostatistics, Düzce University School of Medicine, Düzce, Turkey (MAS)
ORCID IDs of the authors: F.A.A. 0000-0003-2476-768X; Y.D. 0000-0002-2975-2682; H.B.A. 0000-0001-5274-7171; D.C 0000-0001-6597-5632; M.A.S.
0000-0001-5380-0819; İ.Ü. 0000-0002-5649-2257.
Cite this article as: Akcan FA, Dündar Y, Bayram Akcan H, Cebeci D, Sungur MA, Ünlü İ. The Association between Iron Deficiency and Otitis Media
with Effusion. J Int Adv Otol 2019; 15(1): 18-21.
OBJECTIVES: The purpose of this trial is to examine the clinical role of iron metabolism on development of Otitis media with effusion.
MATERIALS METHODS: This prospective study was conducted in a tertiary referral center. The study group made up of children who had surgery
for Otitis media with effusion (OME). Control group was comprised of children who had surgery by a pediatric surgeon for inguinal hernia repair
or circumcision operations with normal ear nose throat examination. Each group was evaluated depending on the serum iron metabolism pa-
rameters.
RESULTS: One-hundred-thirteen children with OME and 117 control patients were included to the study. Iron deficiency anemia was detected in
18 out of 113 patients (15.9%) in study group while there were 4 out of the 117 patients (3.4%) in control group (p:0.001).The mean hemoglobin
level was 12.16 ± 1.16 in OME group and 12.93 ± 1.08 in control group (p<0.001).
CONCLUSION: The current study shows the rate of iron deficiency anemia is higher in patients with OME than controls. Iron-deficiency anemia
might be considered a potential risk factor for development of otitis media with effusion, and iron parameters should be evaluated in these
children.
KEYWORDS: Otitis media with effusion, iron deficiency, anemia, pediatric
INTRODUCTION
Otitis media with effusion (OME) is a disease of middle ear without signs or symptoms of acute suppurative infection [1]. It is a
response of chronic inflammation in the middle ear and the effusion may become persistent in some cases [2]. Despite numerous
therapy options have been tried, there is no current standard therapy for OME [3-5]. If the problem persist for at least 3 months, it
is defined as chronic otitis media with effusion (COME) [6]. Unresolved OME can cause hearing loss, speech delay, and permanent
middle ear damage; if these situations occur then a ventilation tube (VT) should be considered [7, 8]. Many etiologic factors play role
in development of OME but the most important factors are immune system and Eustachian tube dysfunction. Upper airway and
middle ear bacterial or viral infections also play role in OME [9, 10].
Iron metabolism has a critical importance for normal cognitive and motor development in children and furthermore it is essential
for immune system [11, 12]. While many studies in literature show that impaired iron metabolism is associated with chronic inflamma-
tory processes, there is no published study addressing iron metabolism and OME.
This is the first study to investigate the association of iron metabolism as a risk factor in OME. The goal of this study was to analyze
the role of iron deficiency anemia (IDA) on development of OME and also clinical value of IDA.
This study was presented at the “Combined Otolaryngology Spring Meetings (COSM)”, 18-22 April 2018, Maryland, Washington DC, USA.
Corresponding Address: Fatih Alper Akcan E-mail: f.akcan@yahoo.com
Submitted: 13.03.2018 • Revision Received: 28.05.2018 • Accepted: 05.06.2018 • Available Online Date: 27.11.2018
18 ©Copyright 2019 by The European Academy of Otology and Neurotology and The Politzer Society - Available online at www.advancedotology.org
Akcan et al. Risk Factor Assessment
MATERIALS and METHODS Control group was comprised children were examined by pediatric
This prospective study was conducted in a tertiary referral center. The surgeons who confirmed with examination that there was no evi-
local ethics committee approval (Duzce University School of Medi- dence of infection. Children were excluded from the study if patients
cine; 2017/22) and the written informed consent were obtained. The had any of following: history of adenotonsillectomy, craniofacial
study group contained patients who had OME as determined by anomaly, history of VT insertion, chronic diseases such as chronic si-
tympanogram results and physical findings between January 2017 nusitis, allergic rhinitis, asthma bronchial, immunodeficiency or any
and October 2017. Absent light reflex, type B-C tympanogram, glue diagnosed hematologic diseases.
behind the ear drum, air-fluid level on ear drum were defined as signs
of effusion. The study group was followed up for at least three months without
Table 1. Demographics of study and control group. URI: Upper respiratory any medical treatment. At the end of follow-up period, persistent
tract infection cases that were defined as COME. All COME patients underwent
VT placement under general anesthesia. Adenoidectomy was per-
Study Control
formed with VT if the patient was older than 4 years. If there were no
group group
Variable n:113 n:117 p
obstruction findings in patients under 4 years old, only VT was per-
formed. In study group, hemoglobin (Hb), mean corpuscular volume
Age 5.63±2.98 (1-16) 5.67±2.85 (1.5-15) 0.916 (MCV), hematocrit (Hct), ferritin, unsaturated iron binding capacity
Mean siblings 2 (0-4) 2 (1-4) 0.221 (UIBC), serum iron level (SI) were analyzed in addition to routine
blood exams for anesthesia. Iron parameters were also studied from
Sex
control group. Study and control group’ results were compared to the
Male 65 (57.5%) 62 (53.0%) 0.490 World Health Organization’ (WHO) normal range.
Female 48 (42.5%) 55 (47.0%)
Statistical Analysis
Tobacco smoke exposure
Continuous data were abstracted as mean±standard deviation and
Yes 44 (38.9%) 43 (36.8%) 0.733 categorical data were offered as frequency and percent. Categorical
data were analyzed by Pearson chi-square or Fisher’s exact tests de-
No 69 (61.1%) 74 (63.2%)
pending on expected count rule. Comparisons for quantitative vari-
Cow’s milk exposure ables were done using Independent samples t test or Mann-Whitney
before 12 months 38 (33.6%) 44 (37.6%) 0.529 U test depending on distribution of variable, for two groups. Statis-
75 (66.4%) 73 (62.4%) tical analysis was performed with Statistical Package for the Social
Sciences (SPSS) version 22 (IBM Corp.; Armonk, NY, USA) statistical
Breastfeeding at least
6 months only 73 (64.6%) 75 (64.1%) 0.937
package and statistical significance level was considered as 0.05. The
combined effect of age groups and patient / control groups was ex-
40 (35.4%) 42 (35.9%) amined by Two-Way ANOVA.
Frequency of URI during study
period (mean±SD) 216 (1.91±0.92) 172 (1.47±0.65) <0.001 RESULTS
The study population consisted of 65 boys (57.5%) and 48 girls
AOM attack during study
period (mean) 15 (13.3%) 9 (7.7%) 0.166 (42.5%) and the mean age was 5.63±2.98. The control group con-
sisted of 62 boys (53%) and 55 girls (47%) and the mean age was
Hemoglobin<11 g/dL 18 (15.9%) 4 (3.4%) .001 5.67±2.85. Eighteen out of 113 (15.9%) patients had iron deficiency
Ferritin<16 μg/L 32 (28.3%) 11 (9.4%) <0.001 anemia in the study group and 4 out of 117 (3.4%) patients had iron
Statistical significance level was considered as 0.05.
deficiency anemia in the control group (p:0.001). Thirty two out of
AOM: Acute otitis media 113 (28.3%) patients had low ferritin level in the study group and 11
out of 117 (9.4%) patients had low ferritin level in the control group
Table 2. Iron metabolism parameters distribution of study and control group
(p<0.001). The study group (113) had total of 216 upper respirato-
Study Control ry tract infections (URI), while the control group (117) had 172 URIs
group group during the study period. Per-person URI frequency in the study and
Variable n:113 n:117 p control groups were 1.91±0.92 and 1.47±0.65 respectively; and the
Hemoglobin g/dL 12.16±1.16 (9.1-14.8) 12.93±1.08 (9.1-15.5) <0.001 difference was statistically significant (p<0.001). Acute otitis media
(AOM) during the study period was higher in study group but the
Ferritin μg/L 18.20±8.91 (2.9-44.8) 36.97±27.01 (5.0-129.0) <0.001 difference was not significant. Table 1 shows the demographics of all
Serum iron µmol/L 59.74±29.27 (6-133) 68.29±29.14 (18-155) 0.027 groups. There was no significant difference between the groups ad-
dressing to other variables.
UIBC* µmol/L 335.65±68.67 (184-530) 253.21±71.52 (99-430) <0.001
Hct** % 33.12±4.46 38.45±2.17 <0.001 The mean Hb level was 12.16±1.16 in study group and 12.93±1.08 in
MCV*** fL 74.25±10.33 79.94±3.75 0.013 control group (p<0.001). The mean Hct rate was 33.12±4.46 in study
group and 38.45±2.17 in control group (p<0.001). The mean MCV lev-
UIBC: Unsaturated iron binding capacity. Hct: Hematocrit. MCV: Mean corpuscular
volume. el was 74.25±10.33 in study group and 79.94±3.75 in control group
Statistical significance level was considered as 0.05. (p:0.013). The mean ferritin level was 18.20 ± 8.91 in study group and
19
J Int Adv Otol 2019; 15(1): 18-21
36.97±27.01 in control group (p<0.001). The mean serum iron level group but the difference was not statistically significant. The mean
was 59.74±29.27 in study group and 68.29±28.14 in control group hemoglobin levels in IDA and Non-IDA group were 10.51±0.56 and
(p:0.027). The mean UIBC level was 335.65±68.67 in study group and 12.47±0.96 respectively and the difference was statistically signifi-
253.21±71.52 in control group (p<0.001). All parameters were sig- cant (p<0.001) (Table 3).
nificantly different between the groups in terms of iron metabolism.
Table 2 summarizes the iron metabolism parameters distribution of All study and control groups patients were separated into three
study and control group. groups with respect to age. Group-Age interaction was not signifi-
cantly different (p:0.453). The mean Hb levels were significantly
When we analyzed the study group in terms of IDA, we revealed lower in each age group in the study group than the control group
that the Mean Age was significantly lower in IDA group (p:0.004), (p<0.001). In addition, the mean Hb levels between in each age group
were statistically significantly different, for both study and control
Frequency of URI during study period was significantly higher in
groups (p<0.001). Figure 1 summarized the group-age interaction.
IDA group (p:0.010). AOM during the study period was higher in IDA
DISCUSSION
13.5 Study group Iron deficiency is one of the most common nutritional deficiency
Control group
worldwide according to WHO and affects a significant part of the
population [13]. WHO estimates iron deficiency anemia in 2.5% of pe-
13.0
diatric population and describes an abnormality if the rate is more
than 5%. [14] This rate was found as 15.9% in our study group, which is
Hemoglobin (g/dL)
12.5 above the WHO’s estimation of 5%. The evaluation of iron deficiency
starts with measurement of Hb and Hct levels that are non-specific
markers. Ferritin, SI and UIBC are widely used for differential diagno-
12.0 sis and confirmation of iron deficiency. The decreased Hb (<11 g/dL)
and ferritin (<16 μg/L) levels are important for diagnosis of IDA. In our
study, all parameters were significantly different between the groups
11.5
in terms of iron metabolism (Table 2).
11.0 OME is a serious problem in children that can lead to hearing im-
pairment, reduced academic success and impairment in social ad-
≤3 3-≤6 >6
vancement. Even though there are numerous risk factors described
Age groups
for OME, upper respiratory tract infections are among the most im-
Figure 1. Group-Age interaction graphic. The mean hemoglobin levels in each
portant of them. Most of the pathogens that play some role on the
age groups.
development of OME come from the nasopharynx such as Hemophi-
Table 3. Study group distribution in terms of IDA lius influenza, Streptococcus pneumoniae and Moraxella catarrhalis.
Further viral agents, like adenovirus, rhinovirus and influenza play
Study Control
substantial role. Also, Chen et al. [15] showed that, biofilms may play a
group group
Variable n:113 n:117 p major role in the pathogenesis of OME.
Age 3.81±2.41 (1-10) 5.97±2.96 (1-16) 0.004 There are some evidences that IDA may increase susceptibility to
Median siblings 2 (1-4) 2 (0-4) 0.472 infections in pediatric population. This association could be related
with decreased neutrophil and macrophage function, decreased
Sex
production of pro-inflammatory cytokines, and increasing the risk of
Male 12 (66.7%) 53 (55.8%) 0.392 extracellular infections [11, 16]. Iron plays an important role in DNA rep-
Female 6 (33.3%) 42 (44.2%) lication pathways of immunoprogenitor cells [17]. Studies have shown
that iron deficiency is more common in children who have frequent
Tobacco smoke exposure 7 (38.9%) 37 (38.9%) 0.996
upper respiratory tract infections [18, 19]. In the literature, a correlation
Cow’s milk exposure before between low Hb values and increased the infection risk was found
12 months 7 (38.9%) 31 (32.6%) 0.606 when the relationship between anemia and infection was investigat-
Breastfeeding at least ed [18, 19]. Hussain et al. [20] reported that; Hb level <11 g/dL was con-
6 months only 13 (72.2%) 60 (63.2%) 0.461 sidered as low. Mean Hb level was 8.8 g/dL in the study group and
11.6 g/dL in the control group. Anemia rate was found 64.5% in the
Frequency of URI during study
study group and 28.2% in the control group and the anemic group
period (mean±SD) 48 (2.67±1.28) 168 (1.77±0.76) 0.010
was found to be 4.6 times more susceptible to respiratory tract infec-
AOM attack during study period 4 (22.2%) 11 (11.6%) 0.255 tion. Mourad et al. [21] reported that, Anemia was found in 32% of in-
Mean Hemoglobin g/dL 10.51±0.56 12.47±0.96 <0.001 patients and 16% of healthy controls. Mean Hb level was 9.99 ± 0.62
g/dL and 11.99 ± 0.92 g/dL in control group and low Hb level was a
IDA: Iron deficiency anemia. URI: Upper respiratory tract infection. AOM: Acute otitis
media. risk factor for respiratory tract infection. Levy et al. [22] reported that,
Statistical significance level was considered as 0.05. IDA was an independent risk factor for the occurrence of respirato-
20
Akcan et al. Risk Factor Assessment
ry disease (tonsillitis, cold, pneumonia, bronchitis, asthma) and also 2. Daly KA, Hunter LL, Giebink GS. Chronic otitis media with effusion. Pedi-
AOM episodes. In this regard, although there is no clear information atr Rev 1999; 20: 85-93, quiz 94. [CrossRef]
in the literature about which Hb level does indeed cause infections; 3. Rosenfeld RM, Shin JJ, Schwartz SR, Coggins R, Gagnon L, Hackell JM, et
Golz et al. [23] revealed that, 83.8% of children with a Hb value of less al. Clinical practice guideline otitis media with effusion (Update). Otolar-
yngol Head Neck Surg 2016; 154: S1-S41. [CrossRef]
than 9.5 g/dL had a very frequent AOM episode and in conclusion
4. Berkman ND, Wallace IF, Steiner MJ, Harrison M, Greenblatt AM, Lohr
they recommended iron supplementation whenever Hb levels are
KN, et al. Otitis Media with Effusion: Comparative Effectiveness of Treat-
found to be lower than 10 g/dL. In our study. When we analyzed the ments. Comparative Effectiveness Review No. 101. Rockville, MD: Agency
study group in terms of IDA, the mean Hb levels in IDA and Non-IDA for Healthcare Research and Quality. 2013 May. Report No: 13-EHC091-
group were 10.51±0.56 and 12.47±0.96 respectively and the differ- EF.
ence was statistically significant (p<0.001) and the low Hb level was 5. Miura MS, Mascaro M, Rosenfeld RM. Association between otitis media
correlated with the high frequency of URI and AOM (Table 3). Many and gastroesophageal reflux: a systematic review. Otolaryngol Head
investigators identified some risk factors for OME, but iron deficiency Neck Surg 2012; 146: 345-52. [CrossRef]
was never considered to be one of them [2, 3, 5, 9]. In our study the rate 6. Bhutta MF. Epidemiology and pathogenesis of otitis media: construction
of anemia was 15.9% in study group and 3.4% in control group and of a phenotype landscape. Audiol Neurootol 2014; 19: 210-23. [CrossRef]
the difference was statistically significant (p<0.001). 7. Klein JO. The burden of otitis media. Vaccine 2000; 19: S2-8. [CrossRef]
8. Teele DW, Klein JO, Rosner BA. Otitis media with effusion during the first
three years of life and development of speech and language. Pediatrics
The most important limitation of the current study is lack of data
1984; 74: 282-7.
about immunogenic factors such as complements and immunoglob-
9. Kubba H, Pearson JP, Birchall JP. The etiology of otitis media with effu-
ulins. We did not analyze the immunoglobulin and complement lev- sion: a review. Clin Otolaryngol Allied Sci 2000; 25: 181-94. [CrossRef]
els in the study and control groups. However, we excluded patients 10. Ryding M, White P, Kalm O. Eustachian tube function and tympanic
from the study if there were any diagnosed immunologic or hemato- membrane findings after chronic secretory otitis media. Int J Pediatr
logic diseases to avoid these confounding factors. Otorhinolaryngol 2004; 68: 197-204. [CrossRef]
11. Oppenheimer SJ. Iron and its relation to immunity and infectious dis-
This study demonstrates the association of iron deficiency with OME ease. J Nutr 2001; 131: 616-33. [CrossRef]
development and its role of clinical progress. IDA should be evaluat- 12. Halterman JS, Kaczorowski JM, Aligne CA, Auinger P, Szilagyi PG. Iron
ed children with COME which might be a prognostic factor. Despite deficiency and cognitive achievement among school-aged children and
these findings there is no adequate data to determine whether IDA adolescents in the United States. Pediatrics 2001; 107: 1381-6. [CrossRef]
13. Moy RJ. Prevalence, consequences and prevention of childhood nutri-
is cause or effect. The association of IDA with OME requires further
tional iron deficiency: a child public health perspective. Clin Lab Haema-
studies to explain pathophysiology of this association.
tol 2006; 28: 291-8. [CrossRef]
14. WHO. Iron Deficiency Anaemia: Assessment, Prevention and Control. A1.
CONCLUSION Guide for Programme Managers, 2001. Available at:http://www.who.int/
Iron-deficiency anemia might be considered a potential risk factor nutrition/publications/micronutrients/anaemia_iron_deficiency/WHO_
for development of otitis media with effusion, and iron parameters NHD_01.3/en/index.htm (accessed 13.01.2017)
should be evaluated in these children. 15. Chen K, Wu X, Jiang G, Du J, Jiang H. Low dose macrolide administration
for long term is effective for otitis media with effusion in children. Auris
Ethics Committee Approval: Ethics committee approval was received for this Nasus Larynx 2013; 40: 46-50. [CrossRef]
study from the ethic committee of Duzce University (2017/22). 16. Wintergerst ES, Maggini S, Hornig DH. Contribution of selected vitamins
and trace elements to immune function. Ann Nutr Metab 2007; 51: 301-
Informed Consent: Written informed consent was obtained from the patients 23. [CrossRef]
who participated in this study. 17. Muñoz C, Rios E, Olivos J, Brunser O, Olivares M. Iron, copper and immu-
nocompetence. Br J Nutr 2007; 98: 24-28.
Peer-review: Externally peer-reviewed. 18. Busuttil A, Kerr AI, Logan RW. Iron deficiency in children undergoing ad-
enoid-toncillectomy. J Laryngol Otol 1979; 93: 49-58. [CrossRef]
Author Contributions: Concept – F.A.A.; Design - F.A.A., Y.D.; Supervision - 19. Mira E, Benazzo M, Asti L, Marchi A, Spriano P, Losi R. Iron status in chil-
İ.Ü.; Resource - F.A.A., D.C.; Materials - F.A.A., D.C.; Data Collection and/or Pro- dren undergoing toncillectomy and its short-term modification follow-
cessing - F.A.A., D.C.; Analysis and/or Interpretation - H.B.A., M.A.S.; Literature ing surgery. Acta Otolaryngol Suppl 1988; 454: 261-4. [CrossRef]
Search - F.A.A.; Writing - F.A.A., H.B.A.; Critical Reviews - Y.D. 20. Hussain SQ, Ashraf M, Wani JG, Ahmed J. Low Hemoglobin Level a Risk
Factor for Acute Lower Respiratory Tract Infections (ALRTI) in Children. J
Acknowledgements: All authors thank Dr. Dennis Bojrab for his English revision. Clin Diagn Res 2014; 8: 1-3. [CrossRef]
21. Mourad S, Rajab M, Alameddine A, Fares M, Ziade F, Merhi BA. Hemo-
Conflict of Interest: The authors have no conflict of interest to declare. globin level as a risk factor for lower respiratory infections in Lebanese
children. North Am J Med Sci 2010; 2: 461-6. [CrossRef]
Financial Disclosure: The authors declared that this study has received no 22. Levy A, Fraser D, Rosen SD, Dagan R, Deckelbaum RJ, Coles C, et al. Ane-
financial support. mia as a risk factor for infectious diseases in infants and toddlers: results
from a prospective study. Eur J Epidemiol 2005; 20: 277-84. [CrossRef]
REFERENCES 23. Golz A, Netzer A, Goldenberg D, Westerman ST, Joachims HZ. The associ-
1. Tos M. Epidemiology and natural history of secretory otitis. Am J Otol ation between iron-deficiency anemia and recurrent acute otitis media.
1984; 5: 459-62. Am J Otolaryngol 2001; 22: 391-4. [CrossRef]
21
