Skip to main content
Download PDF

Temporal instability of the post-surgical maxillary sinus microbiota

BMC Infectious Diseases volume 18, Article number: 441 (2018) Cite this article

Abstract

Background

Chronic rhinosinusitis is an inflammatory disorder in which the role of bacteria remains uncertain. While sinus outflow obstruction is often an initiating event, mucosal inflammation and dysbiosis may persist or develop in sinuses with widely patent surgical openings. Understanding of the relationship between dysbiosis and chronic sinus inflammation is obfuscated by inter-individual microbiota variability and likely intra-individual temporal variation that has yet to be defined. In this study, long-term microbiota stability is investigated within surgically-opened maxillary sinuses of individuals with and without sinus inflammatory disease.

Methods

Maxillary sinus swabs were performed in 35 subjects with longstanding maxillary antrostomies. Subjects with and without active chronic maxillary sinusitis were included. Repeat swabs were obtained from the same sinuses after a prolonged interval (mean 719 ± 383 days). Patients were categorized based on the inflammatory status of the sinus mucosa at times of sample collection, as assessed by nasal endoscopy. Total DNA from swab eluents was extracted, and the microbiota characterized using 16S rRNA gene sequencing followed by taxonomic classification. Prevalence and abundance of genera were determined by analysis of 16S rRNA gene sequences. Taxa were identified that were stably present between two time points in individual subjects.

Results

The overall proportion of stable taxa across time points was 24.5 ± 10.6%. This stability index was consistent across patient groups and not correlated with clinical parameters. Highly prevalent taxa, including Staphylococcus, Corynebacterium, Propionibacterium, and Pseudomonas, were often stably present, but varied in relative abundance. Janthinobacterium, Enterobacter, Lactobacillus, and Acinetobacter were prevalent and moderately abundant taxa in healthy sinuses, but not in inflamed sinuses. Moraxella and Haemophilus were present at low prevalence and proportional abundance in chronically or intermittently inflamed sinuses, but not in healthy sinuses.

Conclusions

A relatively small component of the post-antrostomy maxillary sinus microbiota exhibits long-term stability in individual subjects. Stable bacteria include a limited number of highly prevalent and a larger number of lower prevalence taxa, which vary widely in proportional abundance. The concept of individual-specific core sinus microbiota, durable over time and medical therapy, but fluctuating in proportional abundance, has implications for understanding the role of bacteria in CRS pathogenesis.

Peer Review reports

Background

Recent interest in the paranasal sinus microbiota, sparked by increased availability of culture-independent methods, has prompted reconsideration of the role of bacteria in chronic rhinosinusitis (CRS). Molecular techniques identify significantly more organisms than standard cultures [1,2,3,4,5], shedding light on complex bacterial communities within healthy sinuses that had previously been assumed to be sterile. The normal microbiota of the sinuses is currently not well defined and appears to vary widely among individuals [2, 6,7,8]. Evidence suggests that diminished bacterial diversity and richness are associated with CRS [2, 5, 6]. It has been proposed that specific organisms may be critical for sinus health and that some bacterial species may be associated with development of disease [2, 9]. At this time, however, it remains to be elucidated whether a stable sinus microbiome exists, and whether disruption of normal microbial homeostasis underlies alterations in host mucosal integrity, mucociliary clearance, or expression of local inflammatory factors. Also unclear is how an individual’s sinonasal microbiome is established and how dynamic it is over time.

The upper airway serves as a filter, constantly entrapping airborne microbes and particulates in the mucus blanket and removing them harmlessly by mucociliary clearance [10, 11]. In settings of mucostasis, overgrowth can occur of bacteria that are either resident or transiently moving through the sinonasal tract. Acute bacterial sinusitis is a common occurrence that is generally self-limited or responsive to medical treatment, suggesting that acute changes in the sinus microbiome associated with mucostasis are tolerated and reversible. While it is possible that pathological alteration of the microbiota precedes inflammation and subsequent mucostasis, evidence for this is lacking in acute sinusitis. In CRS, a more heterogeneous disorder with a multifactorial etiology [12, 13], purulent static secretions are frequently present, and a role for bacteria in promoting chronic inflammation is generally presumed. Antibiotics are the most common medication prescribed for managing symptomatic CRS, but evidence of efficacy is limited, particularly in CRS with nasal polyposis [14].

In cases of medically recalcitrant CRS, endoscopic sinus surgery (ESS) can be very effective in clearing mucopus and enlarging drainage pathways, in order to facilitate mucociliary function and access for topical therapies [15, 16]. The restoration of sinus health after ESS again indicates that changes in the microbiota associated with chronic inflammation are also caused, or at least perpetuated, primarily by mechanical obstruction. In a subset of patients, however, ESS does not reverse the chronic inflammatory process, requiring long-term medical management and sometimes repeated surgery. The role of bacteria in the pathogenesis of chronic sinus mucosal inflammation in this setting remains under investigation [17]. In patients who have undergone adequate ESS, ongoing sinusitis is not due to blockage of sinus ostial outflow. Often, periods of disease control are punctuated by acute inflammatory and/or infectious exacerbations, which can be readily visualized endoscopically and sampled in the office setting. In this way, post-operative sinus cavities present a window through which to investigate the relationship between the microbiome and the mucosal inflammatory state over time.

Previous studies suggest that the sinus microbiome varies among individuals, among sub-sites within an individual sinonasal tract, and in response to medical therapy [18,19,20] or sinus surgery [21,22,23]. While the microbiota remain remarkably constant over time in the gut and urogenital tract [24,25,26], temporal variation has not been evaluated in the post-operative sinuses. In order to evaluate the hypothesis that dysbiosis is a contributing factor in surgically-recalcitrant CRS, it is important to establish what constitutes healthy microbiota and the degree to which a stable set of bacteria, or “core” sinus microbiome, exists for individuals. Only with this understanding can the microbiota characteristics associated with chronic inflammatory sinus disease be interpreted. Therefore, the aim of this study is to determine the long-term stability of the microbiota within surgically-opened maxillary sinuses between two temporally separate points. We hypothesize that stable microbiota features exist within individuals, despite changes in disease state and medical therapy.

Methods

This prospective study was approved by the Johns Hopkins Medicine Institutional Review Board. All subjects provided written consent to participate. Between 3/2011 and 4/2015, 35 patients with longstanding patent maxillary antrostomies, status post ESS in the remote past, had swab samples of the maxillary sinus taken at two distant time points as part of a routine follow-up. All collections were taken from the maxillary sinus mucosa using a cotton-tipped wire swab (Polyester Dacron Swab Pack, MPC, Amarillo, CA; Dry Swabs, Copan Diagnostics, Murrieta, CA) under endoscopic visualization by the same clinician (APL). Swabs were discarded and retaken if contaminated by mucosa outside the target region. Immediately following collection, the samples were flash frozen and stored at −80C until processed.

Definition of groups of patients

The subjects were separated into three groups based on the presence or absence of chronic maxillary sinusitis at the times of sampling, as defined by the EPOS [27] and AAO-HNS guidelines [28] and by endoscopic evidence of inflammation: Group 1 (no CRS): chronic maxillary sinusitis not present at either time point; Group 2 (persistent CRS): persistent chronic maxillary sinusitis at both time points; Group 3 (CRS exacerbation): clinical evidence of chronic maxillary sinusitis exacerbation at one time point only.

In this study design, each subject acts as his/her own control. In other words, one time point can be considered the “baseline” to which the other time point is compared. Additionally, groups 1 and 2, which had the same inflammatory state at both time points, serve as controls for group 3, where different inflammatory states existed at each time point.

The electronic medical records of the patients were reviewed for demographic and clinical information including current sinonasal medications, and recent courses of antibiotics (within 8 weeks of sampling).

DNA extraction

Total genomic DNA was extracted using a protocol developed at the University of Maryland School of Medicine—Institute for Genome Sciences and previously described [29, 30]. Briefly, samples were thawed on ice, incubated in an enzymatic cocktail containing lysozyme, mutanolysin, proteinase K and lysostaphin, after which the microbial cells were lysed using bead beating with silica beads (Lysing Matrix B, MP Biomedicals) with the FastPrep instrument (MBio, Santa Ana, CA). The DNA was then further extracted and purified using the Zymo Fecal DNA kit (Zymo Research, Irvine, CA).

16S rRNA gene PCR amplification and sequencing

The universal primers 319F (ACTCCTACGGGAGGCAGCAG) and 806R (GGACTACHVGGGTWTCTAAT) were used for PCR amplification of the V3-V4 hypervariable regions of the 16S rRNA gene in 96-well microtiter plates using procedures previously published [31]. The 806R primer included a unique 12 base pair (bp) sequence tag to barcode each sample, so that up to 500 samples could be multiplexed in a single Illumina MiSeq (Illumina, San Diego, CA) run. PCR amplifications were performed using Phusion High-Fidelity DNA polymerase (Thermo Fisher Scientific, Waltham, MA) and 10 ng of template DNA in a total reaction volume of 25 μl. Reactions were run in a DNA Engine Tetrad 2 thermo cycler (Bio-Rad, Hercules, CA) using the following cycling parameters: 30 s at 98 °C, followed by 30 cycles of 10 s at 98 °C, 15 s at 66 °C, and 15 s at 72 °C, with a final step of 10 min at 72 °C. The number of PCR cycles chosen is within the range used for standard PCR in molecular evolution and environmental studies [32]. Negative controls without a template were processed for each primer pair. The presence of amplicons was confirmed using gel electrophoresis, after which the SequalPrep Normalization Plate kit (Thermo Fisher Scientific, Waltham, MA) was used for clean-up and normalization (25 ng of 16S PCR amplicon was pooled for each sample), before pooling and 16S rRNA sequencing using the Illumina MiSeq (Illumina, San Diego, CA) (according to the manufacturer’s protocol).

Sequence quality filtering and analysis of 16S rRNA gene sequences

16S rRNA reads were initially screened for low quality bases and short read lengths [33]. We used criteria previously described to assess the quality of sequence reads [34]. Briefly, to pass, a sequence read had to (a) include a perfect match to the sequence tag (barcode) and the 16S rRNA gene primer; (b) be at least 200 bp in length; (c) have no undetermined bases; and (d) have at least a 60% match to a previously determined 16S rRNA gene sequence. Paired-end read pairs were then assembled using PANDAseq [35] and the resulting consensus sequences were de-multiplexed (i.e. assigned to their original sample), trimmed of artificial barcodes and primers, and assessed for chimeras using UCHIME [36] in de novo mode implemented in Quantitative Insights Into Microbial Ecology (QIIME release v. 1.9) [37]. Quality trimmed sequences were then clustered de-novo into Operational Taxonomic Units (OTUs) at 97% similarity cutoff using the uclust algorithm in QIIME, and taxonomic assignments were performed using the Ribosomal Database Project (RDP) classifier implemented in QIIME and the Greengenes database (release 13.8) as a a reference. Singleton sequences were then removed. Estimating and visualization of alpha-diversity indices was performed using R statistical software v.3.2 (Core Team 2013) with packages ggplot218 and Phyloseq v. 1.12.219 [38, 39]. Because alpha diversity metrics can be susceptible to uneven sampling depth between samples, alpha diversity measures were compared after rarefaction to the minimum sampling depth of 500 sequences. When appropriate, taxonomic assignment data were normalized to account for uneven sampling depth with metagenomeSeq’s cumulative sum scaling (CSS; implemented in R) [40].

Statistical analysis

Data were analyzed using SPSS 23.0 (SPSS INC, Chicago, IL). Quantitative data are reported as mean ± SD. The normality of the data distributions was assessed by the Kolmogorov-Smirnov test. One-way ANOVA was used for multiple comparisons between groups, followed by the Scheffé test for post hoc analyses, as appropriate. Linear regression analysis was performed using the least squares method in order to investigate the relationship between different variables. In order to study each individual’s microbiota over time, paired samples from an individual were assessed, and the fraction of shared genera between them was determined, as measured by the binary Jaccard index [41]. For instance, we calculated the fraction of shared genera within a subject between two samples collected hundreds of days apart [Jaccard index (Sample A, Sample B) = SampleA∩SampleB/SampleA∪SampleB]. Microbiota biodiversity metrics included observed OTUs, Shannon diversity index, average proportional abundance defined as the proportional presence of each taxa in a sample, and prevalence defined as the presence or absence of each taxa in a sample. Beta diversity comparisons were performed from CSS-normalized data through principal coordinates analysis plots of Bray-Curtis dissimilarity index determined using QIIME and tested for significance with analysis of similarity (ANOSIM) algorithm (9999 permutations) [42] implemented in the Vegan package in R. A p value of < 0.05 was considered to indicate statistical significance.

Results

Patient demographics

All 35 patients (8 male, 27 female) enrolled in the study and had maxillary sinus samples collected at widely separated time points as described. The demographic characteristic of the whole study group and the three groups separately (group 1: no CRS, group 2: persistent CRS, group 3: CRS exacerbation) are summarized in Tables 1 and 2 respectively. As can be seen, the time interval between sampling was 719 ± 383 days [median (interquartile range) 679 days (490–917 days)]. There were no significant differences in gender, age, or time elapsed between sample collections in the three groups. Groups 2 and 3 had significantly more subjects with nasal polyps than group 1. Group 2 subjects had significantly greater use of oral steroids compared to the other groups, and greater use of nasally-irrigated steroids as compared to group 1. Four patients (four women; two from group 2 and two from group 3) received oral or topical antibiotics during the course of the study. The characteristics of the study cohort without these four patients is presented in Tables S1 and S2 of the Additional file 1.

Table 1 Patients’ characteristics
Full size table
Table 2 Patient characteristics, medication use and microbiome data of the three groups
Full size table

Maxillary sinus microbiome

The 16SrRNA gene sequencing data indicated a total of 314 bacterial genera were detected in the maxillary sinuses of the study group (Additional file 2). Figure 1 shows a heat map representation of the taxa identified in all samples from the three group of patients as well as the Jaccard index of each sample. Across all samples analyzed, Staphylococcus was highly prevalent, identified in 100% of samples, while Pseudomonas was found in 82.7 of samples and Propionibacterium and Corynebacterium were each present in 76.9% of samples. Staphylococcus, Corynebacterium, and Haemophilus were the most abundant genera in post-antrostomy maxillary sinuses. The most prevalent and abundant genera are shown in Tables 3 and 4.

Fig. 1
figure 1

Microbiota of the post-operative maxillary sinus in repeated samples from 35 subjects (patients receiving antibiotics within 8 weeks are noted with red asterisk). In this heatmap visualization, the subjects are divided into groups based on the presence or absence of maxillary sinusitis at the time of sample collection. Each column represents one sample, and subject numbers are listed under paired samples from the same individual. Each row is a bacterial taxon, with the proportional abundance of that taxon depicted by color corresponding to the annotated key (top right). Last row represents Jaccard index of each sample which is depicted by color corresponding to the annotated key (bottom right). The stability index remained similar across groups, suggesting that disease state does not significantly affect this β-diversity metric

Full size image
Table 3 Fifteen most prevalent (left) and proportionally abundant (right) genera in the post-antrostomy maxillary sinus
Full size table
Table 4 Fifteen most prevalent (left) and proportionally abundant (right) genera in the post-antrostomy maxillary sinus
Full size table

For all subjects, the Shannon diversity index was 1.83 ± 0.88. Figure 2 shows observed OTUs and Shannon diversity index in the three groups. Patients of group 2 had significantly lower richness (as expressed by observed OTUs) and diversity compared to the other two groups (Table 2).

Fig. 2
figure 2

Box and whisker plots comparing observed OTUs and diversity (Shannon diversity index) in the three groups of patients. The “box” represents the interquartile range, the thick horizontal bar shows the median value and the “whiskers” represent the minimum and maximum values. Each color represents the paired samples from an individual subject. OTUs: operational taxonomic units; #: p < 0.05

Full size image

Stability of the microbiome over time

In Fig. 3, the microbiota of paired samples are compared, limited to the top 30 genera. The stability of the bacteria present in paired samples from subjects was assessed using the Jaccard index. This demonstrated that an average of 24.5 ± 10.6% of maxillary sinus genera present at one time point are also present at the other. As shown in Table 2, Jaccard index didn’t differ between the three groups (24.2 ± 10.7 for group 1; 25.3 ± 12.7 for group 2; 23.6 ± 11.3 for group 3) Fig. 4 shows the characteristics of the stable maxillary sinus microbiota in the three groups of patients (i.e. taxa present at both time points). Among the highest prevalence organisms, Staphylococcus is stably abundant in all groups, while the proportional abundance of the other highly prevalent taxa, including Propionibacterium, Corynebacterium, Stenotrophomonas, Streptococcus, and Pseudomonas, varies significantly among groups. Several organisms, including Campylobacter and Sphingomonas, are highly prevalent in individual subjects at one time point or the other, but rarely present stably at both times. In healthy sinuses, Janthinobacterium, Enterobacter, Lactobacillus, and Acinetobacter are prevalent and moderately abundant taxa, while these organisms are absent or much less prevalent in inflamed sinuses (Fig. 4). Other bacteria, including Moraxella and Haemophilus, were never found in the stable microbiota of healthy subjects, but were present at low prevalence and proportional abundance in subjects with chronic or intermittent inflammation (Fig. 4).

Fig. 3
figure 3

Relative abundance of top 30 taxa found in repeated samples from the same individual. Group 1: no maxillary sinusitis at either time point; Group 2: persistent maxillary sinusitis at both time points; Group 3: CRS exacerbation at one time point only. In the bar chart, the proportional abundance is denoted by the size of the bar segment, and the identity of the bacterial taxon is denoted by the color, as defined in the key at right

Full size image
Fig. 4
figure 4

a Characteristics of the stable post-antrostomy maxillary sinus; b Graph depicting the relationship between the prevalence of stable taxa and their level of proportional abundance when present. The X-axis shows the prevalence of the taxa at any abundance, and the Y-axis shows the percentage of those subjects in whom the proportional abundance is greater than 1%. Taxa in the upper right are highly prevalent at a high abundance, while taxa in the bottom right are highly prevalent but at relatively low abundance

Full size image

Beta-diversity of nasal microbiota was visualized using ordination. Bray-Curtis dissimilarity index that takes into account the abundance of each OTU (i.e. structure of the microbiome) and estimates of the dissimilarity of the samples was used. ANOSIM tested the hypothesis that samples cluster more closely within patients, rather than between groups. Contrary to our hypothesis, within-patient sample pairs did not cluster together nearly as strongly as they did by group (ANOSIM R = 0.28, p = 0.001; Fig. 5). Figure 5 presents the principal coordinates analysis, where the inflamed time points for group 3 are underlined.

Fig. 5
figure 5

Principal coordinate analysis of Bray–Curtis dissimilarity index generated from taxa of the three groups at both time points. Proportion of variance explained by each principal coordinate axis is denoted in the corresponding axis label. Colored by group and tested with ANOSIM (R = 0.28, p = 0.001). Black circles: group 1; red circles: group 2; green circles: group 3. Inflamed time points for group 3 are underlined

Full size image

Association of stability with clinical parameters

Linear regression analysis did not demonstrate any association between stability index and gender, age, disease state, saline irrigation, topical or systemic steroids, or antibiotics (Table 5).

Table 5 Simple linear regression analysis models for Jaccard index in all patients
Full size table

Discussion

The vast majority of patients with CRS experience clinical resolution after medical and surgical therapy, suggesting that sinus dysbiosis, if present, is either fully reversible or is secondary to impaired ventilation and mucociliary function. In cases of persistent CRS after surgery, the ongoing role of bacteria is unknown. The nose and sinuses are not sterile, and therefore bacteria are necessarily found in both healthy and CRS maxillary sinuses after antrostomy. In this study, we compared the maxillary sinus microbiota in a cohort of patients from whom samples were obtained years apart. Our results confirm that the post-operative maxillary sinus (POMS) microbiota is broadly characterized by a group of highly prevalent taxa which differ widely in proportional abundance. Additionally, there is a diverse sinus microbiota that contains an assortment of proportionally low abundance bacteria that vary between individuals and within individuals over time. The POMS microbiota of an individual person is relatively unstable, with an overlap of approximately 25% of taxa between samples. A slightly higher degree of stability is present in the setting of chronic inflammation where there is a depleted, low diversity microbial community. Additionally, beta-diversity metrics that account for the relative abundance of taxa (Fig. 5), show that sample pairs cluster together more strongly by group than they did within-patient. These findings highlight the complexity of identifying a “normal” or “healthy” microbiome upon which to propose a dysbiotic mechanism responsible for driving CRS. As an individual’s POMS microbiota may be in constant temporal flux, perhaps due to differences in inhaled and skin bacteria transiently entering and being cleared from the sinus, the cause and effect relationship of specific bacteria present in CRS may be difficult to demonstrate. Our findings are in accordance with the results of a recent study, which reported on the middle meatus microbiome of 23 patients undergoing ESS [22]. The authors found a large number of taxa that constitute an extremely small fraction of the total detected microbiota [22].

Previous studies have revealed unique “cores” of bacterial organisms that remains present in samples across various time periods in individual body sites. The gut, which has been studied the most extensively, has a stable core group of bacteria in various proportions. Martinez et al. found 33–40 species-level OTUs that remained consistent over 1 year in three subjects [43]. While this was only 12% of the total OTUs detected, those species accounted for >75% of the total sequences obtained for each subject. Faith et al. studied individual subjects’ fecal microbiota over a 5-year period, showing it to be quite stable with 60% of bacterial strains remaining present [25]. Additionally, the most stable components of the fecal microbiota were also the most abundant. Caporaso et al. [24] extended their study to include the mouth and palms as well as the gut. They too identified a small “core human microbiota” at these body sites present over months, representing about 10% of all OTUs identified at each site [24]. Interestingly, the size of the stable core microbiome was largest for mouth followed by the gut and then the palms respectively. The microbiome of the nasal passages of healthy individuals has been reported to be relatively stable, although a stability index was not calculated [44]. Futhermore, Yan et al. [20] examined the microbiota of the anterior nasal cavity, middle meatus, and sphenoethmoid recess in 12 healthy adults, noting a temporally stable, highly individualized baseline, with some differences in microbiota between the anterior nares and deeper anatomic subsites. Another study that examined middle meatus swabs from 28 healthy subjects found that many individuals harbor respiratory pathogens at low abundance, suggesting that pathogenic bacteria may be transient or permanent members of the healthy microbiome at relatively low amounts [8]. This is consistent with our findings in the POMS.

In the gut, the composition of an individual’s stable microbiota is proposed to be derived from diet or environmental exposures from early in life. In contrast, our findings suggest that the POMS microbiota is much more dynamic. A 25% stability at the genus level may indicate a significant and ongoing influence of transient environmental exposures in the upper airway, which, like the palms, are likely more varied than the gut. At the same time, a number of highly prevalent taxa in the POMS are largely stable in the microbiota but shift in abundance over time. Bacteria belonging to the Staphylococcus, Corynebacterium, Proprionibacterium and Pseudomonas genera are the primary examples. These organisms are often among the most abundant taxa found at any given time point, and, when this is the case, they are present at the other time point in over 80% of subjects.

A limitation of this study is that only two points in time are compared. Obviously, collection of more frequent time points over a greater period would most accurately portray maxillary sinus microbiota temporal variation. We performed a preliminary analysis of additional time points in a smaller number of subjects, which suggested increased microbiota stability over short time intervals, but still with marked fluctuation. As more longitudinal samples are collected and analyzed for individual subjects, characterization of the stably present organisms can be refined. It is likely that medications, dietary influences, and changes in living or work environments lead to shifts in sinus microbiota between sampling points. This is expected and important to recognize in a “real-world” descriptive study such as this. The percentage range of the stable taxa is fairly consistent, despite individual differences in multiple external factors. Indeed, even if we excluded patients who were on antibiotics during sampling or have been on antibiotics in the 2 months preceding sampling, the stability index remains unchanged. The precision to which rare taxa can be identified in the specimens may limit interpretation to some extent. However, based on the depth of coverage, the fact that samples are compared within individuals, and that negative controls without a template were processed for each primer pair during PCR, the possibility of contamination has been minimized as much as possible. Accordingly, we believe that rare taxa were accurately detected and their inclusion is justified in the stability analysis. We observe that very rare taxa present at one time point may be present in high relative abundance at a subsequent time point. This finding highlights the dynamic nature of the sinus microbiota and argues that rare taxa should not be automatically dismissed as irrelevant or contaminants in microbiome studies. Lastly, it should be noted that increased PCR cycle numbers are associated with point mutation artifacts and reduced richness, but do not appear to affect community structure [45].

To the extent that “core” post-antrostomy maxillary sinus microbiota exists, it is likely that these bacteria have the capacity to participate in a “healthy” diverse microbiome as well as to pathologically dominate in a disease setting. The reason for the decreased diversity and evenness associated with CRS is not understood, and a cause-and-effect relationship has not been demonstrated. It is conceivable that changes in the local milieu due to chronic inflammation support the growth of certain resident bacterial species over others, thus driving the community to a less diverse state. Alternatively, development of uneven and less diverse microbiota within the sinus may precede and stimulate a subsequent inflammatory response. Our data suggest that clinically healthy POMS are associated with diverse but relatively temporally unstable microbiota. In cases of longstanding infectious CRS without polyps, a highly depleted microbiota may develop, dominated by few stable taxa. This subset of patients can be frustratingly difficult to rid of inflammation and infection, and this disease phenotype seems most consistent with an etiology linked to dysbiosis. Although long-term control of inflammation is also challenging in CRS with nasal polyps, the association with infection or dysbiosis is not apparent, and the present study shows a similar level of temporal stability to healthy sinuses, with a reduced overall diversity.

Conclusions

Culture independent techniques permit a more comprehensive view of the sinus microbiome than previously available, but interpretation of this information has presented new challenges. Although dysbiosis continues to be hypothesized as contributing to the pathogenesis of CRS, studies to date have not revealed a consistent pattern differentiating CRS patients and healthy controls. In focusing on the post-surgical maxillary sinus, our group has described shifts in the microbiota occurring in response to medical therapy for acute infectious CRS exacerbation, as well as the broad impact of topical therapies [18, 19]. Taken together with the present study of temporal variation, it is clear that the sinus microbiome is complex, variable, and likely influenced constantly by internal and external factors. The identification of temporally stable taxa in uninflamed sinuses may lend further support for their role in maintaining health, but against the larger backdrop, much further investigation is necessary to delineate the relationship between these bacteria and CRS, before consideration can be given to concepts such as probiotic therapy or selection of antibiotics based on culture-independent methods.

Abbreviations

ANOSIM:

Analysis of similarity

bp:

base pair

CRS:

Chronic rhinosinusitis

CSS:

Cumulative sum scaling

ESS:

Endoscopic sinus surgery

OTUs:

Operational taxonomic units

POMS:

Post-operative maxillary sinus

QIIME:

Quantitative insights into microbial ecology

RDP:

Ribosomal database project

References

  1. Aurora R, Chatterjee D, Hentzleman J, Prasad G, Sindwani R, Sanford T. Contrasting the microbiomes from healthy volunteers and patients with chronic rhinosinusitis. JAMA Otolaryngol Head Neck Surg. 2013;139:1328–38.

    Article  PubMed  Google Scholar 

  2. Feazel LM, Robertson CE, Ramakrishnan VR, Frank DN. Microbiome complexity and Staphylococcus aureus in chronic rhinosinusitis. Laryngoscope. 2012;122:467–72.

    Article  PubMed  PubMed Central  Google Scholar 

  3. Hauser LJ, Feazel LM, Ir D, et al. Sinus culture poorly predicts resident microbiota. Int Forum Allergy Rhinol. 2015;5:3–9.

    Article  PubMed  Google Scholar 

  4. Stephenson MF, Mfuna L, Dowd SE, et al. Molecular characterization of the polymicrobial flora in chronic rhinosinusitis. J Otolaryngol Head Neck Surg. 2010;39:182–7.

    PubMed  Google Scholar 

  5. Abreu NA, Nagalingam NA, Song Y, et al. Sinus microbiome diversity depletion and Corynebacterium tuberculostearicum enrichment mediates rhinosinusitis. Sci Transl Med. 2012;4:151ra124.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  6. Biswas K, Hoggard M, Jain R, Taylor MW, Douglas RG. The nasal microbiota in health and disease: variation within and between subjects. Front Microbiol. 2015;9:134.

    Article  PubMed  Google Scholar 

  7. Boase S, Foreman A, Cleland E, et al. The microbiome of chronic rhinosinusitis: culture, molecular diagnostics and biofilm detection. BMC Infect Dis. 2013;13:210.

    Article  PubMed  PubMed Central  Google Scholar 

  8. Ramakrishnan VR, Feazel LM, Gitomer SA, Ir D, Robertson CE, Frank DN. The microbiome of the middle meatus in healthy adults. PLoS One. 2013;8:e85507.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  9. Ramakrishnan VR, Feazel LM, Abrass LJ, Frank DN. Prevalence and abundance of Staphylococcus aureus in the middle meatus of patients with chronic rhinosinusitis, nasal polyps, and asthma. Int Forum Allergy Rhinol. 2013;3:267–71.

    Article  PubMed  Google Scholar 

  10. Geurkink N. Nasal anatomy, physiology, and function. J Allergy Clin Immunol. 1983;72:123–8.

    Article  PubMed  CAS  Google Scholar 

  11. Jones N. The nose and paranasal sinuses physiology and anatomy. Adv Drug Deliv Rev. 2001;51:5–19.

    Article  PubMed  CAS  Google Scholar 

  12. Tomassen P, Van Zele T, Zhang N, et al. Pathophysiology of chronic rhinosinusitis. Proc Am Thorac Soc. 2011;8:115–20.

    Article  PubMed  CAS  Google Scholar 

  13. Van Crombruggen K, Zhang N, Gevaert P, Tomassen P, Bachert C. Pathogenesis of chronic rhinosinusitis: inflammation. J Allergy Clin Immunol. 2011;128:728–32.

    Article  PubMed  CAS  Google Scholar 

  14. Bhattacharyya N. Antimicrobial therapy in chronic rhinosinusitis. Curr Allergy Asthma Rep. 2009;9:221–6.

    Article  PubMed  CAS  Google Scholar 

  15. Harvey RJ, Goddard JC, Wise SK, Schlosser RJ. Effects of endoscopic sinus surgery and delivery device on cadaver sinus irrigation. Otolaryngol Head Neck Surg. 2008;139:137–42.

    Article  PubMed  Google Scholar 

  16. Smith KA, Smith TL, Mace JC, Rudmik L. Endoscopic sinus surgery compared to continued medical therapy for patients with refractory chronic rhinosinusitis. Int Forum Allergy Rhinol. 2014;4:823–7.

    Article  PubMed  PubMed Central  Google Scholar 

  17. Chalermwatanachai T, Zhang N, Holtappels G, Bachert C. Association of Mucosal Organisms with patterns of inflammation in chronic Rhinosinusitis. PLoS One. 2015;10:e0136068.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  18. Liu CM, Kohanski MA, Mendiola M, et al. Impact of saline irrigation and topical corticosteroids on the postsurgical sinonasal microbiota. Int Forum Allergy Rhinol. 2015;5:185–90.

    Article  PubMed  Google Scholar 

  19. Liu CM, Soldanova K, Nordstrom L, et al. Medical therapy reduces microbiota diversity and evenness in surgically recalcitrant chronic rhinosinusitis. Int Forum Allergy Rhinol. 2013;3:775–81.

    Article  PubMed  PubMed Central  Google Scholar 

  20. Yan M, Pamp SJ, Fukuyama J, et al. Nasal microenvironments and interspecific interactions influence nasal microbiota complexity and S. aureus carriage. Cell Host Microbe. 2013;14:631–40.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  21. Cleland EJ, Bassiouni A, Vreugde S, Wormald PJ. The bacterial microbiome in chronic rhinosinusitis: richness, diversity, postoperative changes, and patient outcomes. Am J Rhinol Allergy. 2016;30:37–43.

    Article  PubMed  Google Scholar 

  22. Jain R, Hoggard M, Biswas K, Zoing M, Jiang Y, Douglas R. Changes in the bacterial microbiome of patients with chronic rhinosinusitis after endoscopic sinus surgery. Int Forum Allergy Rhinol. 2017;7:7–15.

    Article  PubMed  Google Scholar 

  23. Hauser LJ, Ir D, Kingdom TT, Robertson CE, Frank DN, Ramakrishnan VR. Investigation of bacterial repopulation after sinus surgery and perioperative antibiotics. Int Forum Allergy Rhinol. 2016;6:34–40.

    Article  PubMed  Google Scholar 

  24. Caporaso JG, Lauber CL, Costello EK, et al. Moving pictures of the human microbiome. Genome Biol. 2011;12:R50.

    Article  PubMed  PubMed Central  Google Scholar 

  25. Faith JJ, Guruge JL, Charbonneau M, et al. The long-term stability of the human gut microbiota. Science. 2013;341:1237439.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  26. Li K, Bihan M, Methe BA. Analyses of the stability and core taxonomic memberships of the human microbiome. PLoS One. 2013;8:e63139.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  27. Fokkens WJ, Lund VJ, Mullol J, et al. EPOS 2012: European position paper on rhinosinusitis and nasal polyps 2012. A summary for otorhinolaryngologists. Rhinology. 2012;50:1–12.

    Article  PubMed  Google Scholar 

  28. Rosenfeld RM, Piccirillo JF, Chandrasekhar SS, et al. Clinical practice guideline (update): Adult Sinusitis Executive Summary. Otolaryngol Head Neck Surg. 2015;152:598–609.

    Article  PubMed  Google Scholar 

  29. Ravel J, Gajer P, Abdo Z, et al. Vaginal microbiome of reproductive-age women. Proc Natl Acad Sci U S A. 2011;108(Suppl 1):4680–7.

    Article  PubMed  Google Scholar 

  30. Zupancic ML, Cantarel BL, Liu Z, et al. Analysis of the gut microbiota in the old order Amish and its relation to the metabolic syndrome. PLoS One. 2012;7:e43052.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  31. Roghmann MC, Lydecker AD, Hittle L, DeBoy RT, Nowak RG, Johnson JC, Mongodin EF. Comparison of the microbiota of older adults living in nursing homes and the community. mSphere. 2017;2:e00210–7.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  32. Galloway-Peña J, Smith D, Sahasrabhojane P, et al. Characterization of oral and gut microbiome temporal variability in hospitalized cancer patients. Genome Med. 2017;9:21.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  33. Fadrosh DW, Ma B, Gajer P, et al. An improved dual-indexing approach for multiplexed 16S rRNA gene sequencing on the Illumina MiSeq platform. Microbiome. 2014;2:6.

    Article  PubMed  PubMed Central  Google Scholar 

  34. Sellitto M, Bai G, Serena G, et al. Proof of concept of microbiome-metabolome analysis and delayed gluten exposure on celiac disease autoimmunity in genetically at-risk infants. PLoS One. 2012;7:e33387.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  35. Masella AP, Bartram AK, Truszkowski JM, Brown DG, Neufeld JD. PANDAseq: paired-end assembler for illumina sequences. BMC Bioinformatics. 2012;13:31.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  36. Edgar RC, Haas BJ, Clemente JC, Quince C, Knight R. UCHIME improves sensitivity and speed of chimera detection. Bioinformatics. 2011;27:2194–200.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  37. Caporaso JG, Kuczynski J, Stombaugh J, et al. QIIME allows analysis of high-throughput community sequencing data. Nat Methods. 2010;7:335–6.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  38. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. 2013; http://www.R-project.org. Accessed on April 2016.

  39. McMurdie PJ, Holmes S. Phyloseq: a bioconductor package for handling and analysis of high-throughput phylogenetic sequence data. Pac Symp Biocomput. 2012:235–46. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3357092/pdf/nihms-374670.pdf.

  40. Paulson JN, Stine OC, Bravo HC, Pop M. Differential abundance analysis for microbial marker-gene surveys. Nat Methods. 2013;10:1200–2.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  41. Jaccard P. The distribution of the flora in the alpine zone. New Phytol. 1912;11:37–50.

    Article  Google Scholar 

  42. Clarke K, Warwick R. Change in marine communities: an approach to statistical analysis and interpretation. Plymouth: Natural Environment Research Council; 1994.

    Google Scholar 

  43. Martinez I, Muller CE, Walter J. Long-term temporal analysis of the human fecal microbiota revealed a stable core of dominant bacterial species. PLoS One. 2013;8:e69621.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  44. Frank DN, Feazel LM, Bessessen MT, et al. The human nasal microbiotia and Staphylococcus aureus colonization. PLoS One. 2010;5:e10598.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  45. Sipos R, Székely AJ, Palatinszky M, Révész S, Márialigeti K, Nikolausz M. Effect of primer mismatch, annealing temperature and PCR cycle number on 16S rRNA gene-targetting bacterial community analysis. FEMS Microbiol Ecol. 2007;60:341–50.

    Article  PubMed  CAS  Google Scholar 

Download references

Funding

Funding for this work was provided by NIH (R15DE021194 and R01 AI072502) to APL.

Availability of data and materials

The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.

Author information

Authors and Affiliations

  1. Department of Otolaryngology – Head and Neck Surgery, Johns Hopkins University School of Medicine, Johns Hopkins Outpatient Center, 6th floor, 601 N. Caroline Street, Baltimore, MD, 21287-0910, USA

    Ioannis Koutsourelakis, Ashleigh Halderman, Syed Khalil, Lauren E. Hittle & Andrew P. Lane

  2. Institute for Genome Sciences, Department of Microbiology & Immunology, University of Maryland School of Medicine, Baltimore, USA

    Emmanuel F. Mongodin

Authors
  1. Ioannis Koutsourelakis
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Ashleigh Halderman
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Syed Khalil
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Lauren E. Hittle
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Emmanuel F. Mongodin
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Andrew P. Lane
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

APL contributed to the project conception, design, data analysis, data acquisition, drafting of the manuscript. IK contributed to the data analysis, data acquisition, and drafting of the manuscript. AH contributed to the project conception, design, data analysis, data acquisition, and drafting of the manuscript. SK and LEH contributed to the project design, data analysis, and data acquisition. EFM contributed to the data analysis, and drafting of the manuscript. All authors have read and approved the manuscript.

Corresponding author

Correspondence to Andrew P. Lane.

Ethics declarations

Ethics approval and consent to participate

All subjects provided written consent to participate. The study was approved by the Johns Hopkins Medicine Institutional Review Board.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Additional files

Additional file 1:

Table S1. Patients’ characteristics (excluding patients receiving antibiotics with 8 weeks of sample collection). Table S2. Patient characteristics, medication use and microbiome data of the three groups (excluding patients receiving antibiotics with 8 weeks of sample collection). (DOC 39 kb)

Additional file 2:

16S rRNA gene sequences of all samples presented by group. (XLS 255 kb)

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

Reprints and Permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Koutsourelakis, I., Halderman, A., Khalil, S. et al. Temporal instability of the post-surgical maxillary sinus microbiota. BMC Infect Dis 18, 441 (2018). https://doi.org/10.1186/s12879-018-3272-9

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12879-018-3272-9

Keywords

BMC Infectious Diseases

ISSN: 1471-2334

Contact us