506257 research-article2013
CPJ521210.1177/0009922813506257Clinical PediatricsShaikh et al
Article
Are Nasopharyngeal Cultures Useful in Diagnosis of Acute Bacterial Sinusitis in Children?
Clinical Pediatrics 52(12) 1118–1121 © The Author(s) 2013 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/0009922813506257 cpj.sagepub.com
Nader Shaikh, MD, MPH1, Alejandro Hoberman, MD1, D. Kathleen Colborn, BS1, Diana H. Kearney, RN, CCRC1, Jong H. Jeong, PhD3, Marcia Kurs-Lasky, MS1, Karen A. Barbadora, MT4, A’Delbert Bowen, MD4, Lynda L. Flom, MD4, and Ellen R. Wald, MD2
Abstract The diagnosis of acute bacterial sinusitis can be challenging because symptoms of acute sinusitis and an upper respiratory tract infection (URI) overlap. A rapid test, if accurate in differentiating sinusitis from URI, could be helpful in the diagnostic process. We examined the utility of nasopharyngeal cultures in identifying the subgroup of children with a clinical diagnosis of acute sinusitis who are least likely to benefit from antimicrobial therapy (those with completely normal sinus radiographs). Nasopharyngeal swabs were collected from 204 children meeting a priori clinical criteria for acute sinusitis. All children had sinus X-rays at the time of diagnosis. To determine if negative nasopharyngeal culture results could reliably identify the subgroup of children with normal radiographs, we calculated negative predictive values and negative likelihood ratios. Absence of pathogens in the nasopharynx was not helpful in identifying this low-risk subgroup. Keywords pediatrics, nasopharyngeal, flora, colonization, symptoms
Background Acute bacterial sinusitis is a complication of a viral upper respiratory tract infection (URI). Inflammation of the mucosal lining of the nose and sinuses secondary to the viral infection sets the stage for bacterial superinfection. The microorganisms that infect the paranasal sinuses originate in the nasopharynx. Thus, on a theoretical level, the presence of pathogens in the nasopharynx and the development of bacterial sinusitis are closely related. Because symptoms of acute sinusitis and URI overlap, the diagnosis of sinusitis based on clinical findings alone can be challenging. Approximately 20% of children meeting stringent clinical criteria for sinusitis have completely normal sinus radiographs,1 suggesting that clinical criteria alone may lack specificity. Accordingly, a rapid test, if accurate in differentiating true sinusitis from URI, could be helpful in reducing unnecessary use of antibiotics. Specifically, if the lack of pathogens colonizing the nasopharynx was closely associated with having normal sinus radiographs, then antimicrobial therapy could be limited to children with colonization. In acute otitis media, which has a similar anatomy, pathophysiology, and microbiology
as sinusitis, lack of nasopharyngeal colonization with Streptococcus pneumoniae, Haemophilus influenzae, and Moraxella catarrhalis predicts a negative middle ear fluid culture.2,3 Accordingly, testing for the presence of pathogens in the nasopharynx, especially if rapid antigen testing or polymerase chain reaction–based kits become available, could conceivably play a role in identifying children with bacterial disease or in contrast, the absence of colonization of the nasopharynx might identify children at very low or no risk for bacterial sinusitis. The objective of this study was to determine whether, in children meeting clinical criteria for sinusitis, nasopharyngeal cultures could accurately identify children 1
University of Pittsburgh School of Medicine, Pittsburgh, PA, USA University of Wisconsin School of Medicine and Public Health, Madison, WI, USA 3 University of Pittsburgh Graduate School of Public Health, Pittsburgh, PA, USA 4 Children’s Hospital of Pittsburgh, Pittsburgh, PA, USA 2
Corresponding Author: Nader Shaikh, General Academic Pediatrics, Children’s Hospital of Pittsburgh, 4401 Penn Avenue, Pittsburgh, PA 15224, USA. Email: [email protected]
Downloaded from cpj.sagepub.com at Univ of Connecticut / Health Center / Library on April 12, 2015
1119
Shaikh et al with completely normal sinus radiographs. Although plain radiographs are neither necessary nor desirable for the diagnosis of sinusitis, children with completely normal radiographs are unlikely to have significant bacterial disease. In other words, even though it has been shown that positive radiographs are not at all useful for confirming a diagnosis of sinusitis,4 negative X-rays are useful in ruling out sinusitis.5 Furthermore, no other reference standard is available for the diagnosis of acute bacterial sinusitis; unlike tympanocentesis, needle aspiration of the sinuses is an impractical procedure in the primary care setting.
Methods We prospectively enrolled consecutive children aged 2 to 12 years presenting to 1 of 4 general ambulatory pediatric clinics in and around Pittsburgh (2 urban, 2 suburban) during 2 consecutive respiratory seasons. Most (90%) children were enrolled during the winter months (October to March of 2008-2009 and 20092010). Children meeting stringently defined a priori clinical criteria consistent with the latest American Academy of Pediatrics guidelines on sinusitis were considered eligible.6 Namely, we included children with (a) persistent URI symptoms (ie, 10-29 days of cough [not exclusively nocturnal] and/or nasal symptoms [rhinorrhea of any quality]) who were not improving or (b) worsening symptoms (substantial worsening of nasal symptoms and/or fever after a period of improvement). We excluded children who had received antimicrobial treatment within 7 days before presentation, had evidence of another infection (ie, acute otitis media or pneumonia), or who had underlying immune deficiency, cystic fibrosis, ciliary dyskinesis, or major developmental delay. Children with asthma were included only if they were not wheezing on exam and if they had nasal symptoms that were worsening or persistent. Children with a history of allergic rhinitis were included only if their respiratory symptoms had acutely worsened. This study was completed well after immunization with the 7-valent pneumococcal was introduced but before immunization with the 13-valent pneumococcal vaccination had started. We obtained nasopharyngeal cultures at the baseline visit from one nostril using a sterile, flexible, thin, polyester/rayon-tipped moistened swab. The tip of the nose was raised and the swab introduced gently along the floor of the nasal cavity, passing under the inferior turbinate until the pharyngeal wall was reached. The swab was placed into Amies transport media (without charcoal), refrigerated, transported to the lab for culturing, and inoculated onto a trypticase soy 5% sheep blood
agar and a chocolate plate. Cultures were incubated overnight at 37°C with 5% CO2. If no growth was present after overnight incubation, the culture was reincubated for another 24 hours. Growth of pathogenic bacteria (S pneumoniae, H influenzae, and M catarrhalis) was assessed using standard semiquantitative techniques (growth in 0, 1, 2, 3, or 4 quadrants); 3+ and 4+ were considered heavy growth. Staphylococcus aureus was not considered a pathogen. In some cases, the nasopharyngeal culture was not obtained because of investigator oversight or parental time constraints. We obtained anteroposterior and occipitomental (Water’s view) radiographs on all children in the study on the day of diagnosis. We also obtained lateral views for children 6 years or older to permit assessment of the frontal and sphenoid sinuses (both frontal and sphenoid sinusitis are unusual in children younger than 6 years). Sinus radiographs were independently reviewed by 2 radiologists (LLF, AB) who were unaware of any clinical data. Disagreements between the 2 radiologists were resolved by discussion. Previous studies have used mucosal thickening of ≥4 mm to indicate the presence of sinusitis. However, it is now clear that no degree of thickening (nor complete opacification for that matter) is sufficient to rule in sinusitis; many children with uncomplicated viral URI will have abnormal sinus X-rays. In addition, because the ethmoidal air cells are superimposed on each other on X-ray, the degree of mucosal thickening cannot be reliably determined. In contrast, a radiograph in which all sinuses are completely aerated is a very good indicator that the child’s respiratory symptoms do not represent an episode of significant sinus disease. Thus, instead of using the threshold of 4 mm (that was developed to rule in disease), we used the absence of any mucosal thickening (to indicate absence of significant sinus disease). We calculated positive and negative predictive values and likelihood ratios according to the type of bacterial species.7,8 Because we were specifically interested in whether negative nasopharyngeal culture results could reliably identify those with normal radiographs, our focus was on the negative predictive values and negative likelihood ratios. We considered findings significant when the confidence interval for the likelihood ratio did not include 1.0. We considered a finding clinically significant if the negative likelihood ratio exceeded 2.0.
Results Of the 258 children meeting a priori clinical criteria for acute sinusitis, 204 had nasopharyngeal swabs performed. Nasopharyngeal specimens were not collected in 54 children (mostly because of parental refusal); these children were followed as part of a larger longitudinal
Downloaded from cpj.sagepub.com at Univ of Connecticut / Health Center / Library on April 12, 2015
1120
Clinical Pediatrics 52(12)
Table 1. Likelihood of Bacteriologic Findings to Predict Normal Sinus X-Rays. Predictive Value, % (95% Confidence Interval)
X-Ray Results (n) Bacterial Species Recovered From the Nasopharynx Streptococcus pneumoniae Streptococcus pneumoniae, heavy growth Haemophilus influenzae Haemophilus influenzae, heavy growth Moraxella catarrhalis Moraxella catarrhalis, heavy growth No pathogens Pathogen(s), ≥1 Pathogen(s), ≥2
Likelihood Ratio (95% Confidence Interval)
Normal (n = 41)
Abnormal (n = 163)
Positive Predictive Value
Negative Predictive Value
Positive Likelihood Ratio
Negative Likelihood Ratio
10 2
52 42
16.1 (8.0-27.7) 4.6 (0.6-15.5)
78.2 (70.5-84.7) 75.6 (68.2-82.1)
0.76 (0.43-1.37) 0.19 (0.05-0.75)
1.11 (0.91-1.36) 1.28 (1.14-1.44)
4 2 7 3 24 17 4
27 19 19 17 84 79 18
12.9 (3.6-29.8) 9.50 (1.2-30.4) 26.9 (11.6-47.8) 15.0 (3.2-37.9) 22.2 (14.8-31.2) 17.7 (10.7-26.8) 18.2 (5.2-40.3)
78.6 (71.8-84.5) 78.7 (72.0-84.4) 80.9 (74.3-86.4) 79.4 (72.8-85.0) 82.3 (73.2-89.3) 77.8 (68.8-85.2) 79.7 (73.1-85.3)
0.59 (0.22-1.59) 0.42 (0.10-1.72) 1.46 (0.66-3.25) 0.70 (0.22-2.28) 1.14 (0.84-1.53) 0.86 (0.58-1.27) 0.88 (0.32-2.47)
1.08 (0.96-1.22) 1.08 (0.99-1.18) 0.94 (0.81-1.09) 1.03 (0.94-1.14) 0.86 (0.58-1.27) 1.14 (0.84-1.53) 1.01 (0.91-1.14)
study, but are not included within the current data set. Parents in the suburban clinics were significantly more likely to refuse the nasopharyngeal swab than the parents in the inner-city urban clinics. As such, children with missing cultures (who are not included in the study) were more likely to be Caucasian. There were no significant differences in age or gender between children with and without nasopharyngeal cultures. The mean age (standard deviation) of the 204 children included in the present analysis was 6.2 (2.8) years. Younger children were significantly more likely to be colonized with S pneumoniae (38.4% of children
CPJ521210.1177/0009922813506257Clinical PediatricsShaikh et al
Article
Are Nasopharyngeal Cultures Useful in Diagnosis of Acute Bacterial Sinusitis in Children?
Clinical Pediatrics 52(12) 1118–1121 © The Author(s) 2013 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/0009922813506257 cpj.sagepub.com
Nader Shaikh, MD, MPH1, Alejandro Hoberman, MD1, D. Kathleen Colborn, BS1, Diana H. Kearney, RN, CCRC1, Jong H. Jeong, PhD3, Marcia Kurs-Lasky, MS1, Karen A. Barbadora, MT4, A’Delbert Bowen, MD4, Lynda L. Flom, MD4, and Ellen R. Wald, MD2
Abstract The diagnosis of acute bacterial sinusitis can be challenging because symptoms of acute sinusitis and an upper respiratory tract infection (URI) overlap. A rapid test, if accurate in differentiating sinusitis from URI, could be helpful in the diagnostic process. We examined the utility of nasopharyngeal cultures in identifying the subgroup of children with a clinical diagnosis of acute sinusitis who are least likely to benefit from antimicrobial therapy (those with completely normal sinus radiographs). Nasopharyngeal swabs were collected from 204 children meeting a priori clinical criteria for acute sinusitis. All children had sinus X-rays at the time of diagnosis. To determine if negative nasopharyngeal culture results could reliably identify the subgroup of children with normal radiographs, we calculated negative predictive values and negative likelihood ratios. Absence of pathogens in the nasopharynx was not helpful in identifying this low-risk subgroup. Keywords pediatrics, nasopharyngeal, flora, colonization, symptoms
Background Acute bacterial sinusitis is a complication of a viral upper respiratory tract infection (URI). Inflammation of the mucosal lining of the nose and sinuses secondary to the viral infection sets the stage for bacterial superinfection. The microorganisms that infect the paranasal sinuses originate in the nasopharynx. Thus, on a theoretical level, the presence of pathogens in the nasopharynx and the development of bacterial sinusitis are closely related. Because symptoms of acute sinusitis and URI overlap, the diagnosis of sinusitis based on clinical findings alone can be challenging. Approximately 20% of children meeting stringent clinical criteria for sinusitis have completely normal sinus radiographs,1 suggesting that clinical criteria alone may lack specificity. Accordingly, a rapid test, if accurate in differentiating true sinusitis from URI, could be helpful in reducing unnecessary use of antibiotics. Specifically, if the lack of pathogens colonizing the nasopharynx was closely associated with having normal sinus radiographs, then antimicrobial therapy could be limited to children with colonization. In acute otitis media, which has a similar anatomy, pathophysiology, and microbiology
as sinusitis, lack of nasopharyngeal colonization with Streptococcus pneumoniae, Haemophilus influenzae, and Moraxella catarrhalis predicts a negative middle ear fluid culture.2,3 Accordingly, testing for the presence of pathogens in the nasopharynx, especially if rapid antigen testing or polymerase chain reaction–based kits become available, could conceivably play a role in identifying children with bacterial disease or in contrast, the absence of colonization of the nasopharynx might identify children at very low or no risk for bacterial sinusitis. The objective of this study was to determine whether, in children meeting clinical criteria for sinusitis, nasopharyngeal cultures could accurately identify children 1
University of Pittsburgh School of Medicine, Pittsburgh, PA, USA University of Wisconsin School of Medicine and Public Health, Madison, WI, USA 3 University of Pittsburgh Graduate School of Public Health, Pittsburgh, PA, USA 4 Children’s Hospital of Pittsburgh, Pittsburgh, PA, USA 2
Corresponding Author: Nader Shaikh, General Academic Pediatrics, Children’s Hospital of Pittsburgh, 4401 Penn Avenue, Pittsburgh, PA 15224, USA. Email: [email protected]
Downloaded from cpj.sagepub.com at Univ of Connecticut / Health Center / Library on April 12, 2015
1119
Shaikh et al with completely normal sinus radiographs. Although plain radiographs are neither necessary nor desirable for the diagnosis of sinusitis, children with completely normal radiographs are unlikely to have significant bacterial disease. In other words, even though it has been shown that positive radiographs are not at all useful for confirming a diagnosis of sinusitis,4 negative X-rays are useful in ruling out sinusitis.5 Furthermore, no other reference standard is available for the diagnosis of acute bacterial sinusitis; unlike tympanocentesis, needle aspiration of the sinuses is an impractical procedure in the primary care setting.
Methods We prospectively enrolled consecutive children aged 2 to 12 years presenting to 1 of 4 general ambulatory pediatric clinics in and around Pittsburgh (2 urban, 2 suburban) during 2 consecutive respiratory seasons. Most (90%) children were enrolled during the winter months (October to March of 2008-2009 and 20092010). Children meeting stringently defined a priori clinical criteria consistent with the latest American Academy of Pediatrics guidelines on sinusitis were considered eligible.6 Namely, we included children with (a) persistent URI symptoms (ie, 10-29 days of cough [not exclusively nocturnal] and/or nasal symptoms [rhinorrhea of any quality]) who were not improving or (b) worsening symptoms (substantial worsening of nasal symptoms and/or fever after a period of improvement). We excluded children who had received antimicrobial treatment within 7 days before presentation, had evidence of another infection (ie, acute otitis media or pneumonia), or who had underlying immune deficiency, cystic fibrosis, ciliary dyskinesis, or major developmental delay. Children with asthma were included only if they were not wheezing on exam and if they had nasal symptoms that were worsening or persistent. Children with a history of allergic rhinitis were included only if their respiratory symptoms had acutely worsened. This study was completed well after immunization with the 7-valent pneumococcal was introduced but before immunization with the 13-valent pneumococcal vaccination had started. We obtained nasopharyngeal cultures at the baseline visit from one nostril using a sterile, flexible, thin, polyester/rayon-tipped moistened swab. The tip of the nose was raised and the swab introduced gently along the floor of the nasal cavity, passing under the inferior turbinate until the pharyngeal wall was reached. The swab was placed into Amies transport media (without charcoal), refrigerated, transported to the lab for culturing, and inoculated onto a trypticase soy 5% sheep blood
agar and a chocolate plate. Cultures were incubated overnight at 37°C with 5% CO2. If no growth was present after overnight incubation, the culture was reincubated for another 24 hours. Growth of pathogenic bacteria (S pneumoniae, H influenzae, and M catarrhalis) was assessed using standard semiquantitative techniques (growth in 0, 1, 2, 3, or 4 quadrants); 3+ and 4+ were considered heavy growth. Staphylococcus aureus was not considered a pathogen. In some cases, the nasopharyngeal culture was not obtained because of investigator oversight or parental time constraints. We obtained anteroposterior and occipitomental (Water’s view) radiographs on all children in the study on the day of diagnosis. We also obtained lateral views for children 6 years or older to permit assessment of the frontal and sphenoid sinuses (both frontal and sphenoid sinusitis are unusual in children younger than 6 years). Sinus radiographs were independently reviewed by 2 radiologists (LLF, AB) who were unaware of any clinical data. Disagreements between the 2 radiologists were resolved by discussion. Previous studies have used mucosal thickening of ≥4 mm to indicate the presence of sinusitis. However, it is now clear that no degree of thickening (nor complete opacification for that matter) is sufficient to rule in sinusitis; many children with uncomplicated viral URI will have abnormal sinus X-rays. In addition, because the ethmoidal air cells are superimposed on each other on X-ray, the degree of mucosal thickening cannot be reliably determined. In contrast, a radiograph in which all sinuses are completely aerated is a very good indicator that the child’s respiratory symptoms do not represent an episode of significant sinus disease. Thus, instead of using the threshold of 4 mm (that was developed to rule in disease), we used the absence of any mucosal thickening (to indicate absence of significant sinus disease). We calculated positive and negative predictive values and likelihood ratios according to the type of bacterial species.7,8 Because we were specifically interested in whether negative nasopharyngeal culture results could reliably identify those with normal radiographs, our focus was on the negative predictive values and negative likelihood ratios. We considered findings significant when the confidence interval for the likelihood ratio did not include 1.0. We considered a finding clinically significant if the negative likelihood ratio exceeded 2.0.
Results Of the 258 children meeting a priori clinical criteria for acute sinusitis, 204 had nasopharyngeal swabs performed. Nasopharyngeal specimens were not collected in 54 children (mostly because of parental refusal); these children were followed as part of a larger longitudinal
Downloaded from cpj.sagepub.com at Univ of Connecticut / Health Center / Library on April 12, 2015
1120
Clinical Pediatrics 52(12)
Table 1. Likelihood of Bacteriologic Findings to Predict Normal Sinus X-Rays. Predictive Value, % (95% Confidence Interval)
X-Ray Results (n) Bacterial Species Recovered From the Nasopharynx Streptococcus pneumoniae Streptococcus pneumoniae, heavy growth Haemophilus influenzae Haemophilus influenzae, heavy growth Moraxella catarrhalis Moraxella catarrhalis, heavy growth No pathogens Pathogen(s), ≥1 Pathogen(s), ≥2
Likelihood Ratio (95% Confidence Interval)
Normal (n = 41)
Abnormal (n = 163)
Positive Predictive Value
Negative Predictive Value
Positive Likelihood Ratio
Negative Likelihood Ratio
10 2
52 42
16.1 (8.0-27.7) 4.6 (0.6-15.5)
78.2 (70.5-84.7) 75.6 (68.2-82.1)
0.76 (0.43-1.37) 0.19 (0.05-0.75)
1.11 (0.91-1.36) 1.28 (1.14-1.44)
4 2 7 3 24 17 4
27 19 19 17 84 79 18
12.9 (3.6-29.8) 9.50 (1.2-30.4) 26.9 (11.6-47.8) 15.0 (3.2-37.9) 22.2 (14.8-31.2) 17.7 (10.7-26.8) 18.2 (5.2-40.3)
78.6 (71.8-84.5) 78.7 (72.0-84.4) 80.9 (74.3-86.4) 79.4 (72.8-85.0) 82.3 (73.2-89.3) 77.8 (68.8-85.2) 79.7 (73.1-85.3)
0.59 (0.22-1.59) 0.42 (0.10-1.72) 1.46 (0.66-3.25) 0.70 (0.22-2.28) 1.14 (0.84-1.53) 0.86 (0.58-1.27) 0.88 (0.32-2.47)
1.08 (0.96-1.22) 1.08 (0.99-1.18) 0.94 (0.81-1.09) 1.03 (0.94-1.14) 0.86 (0.58-1.27) 1.14 (0.84-1.53) 1.01 (0.91-1.14)
study, but are not included within the current data set. Parents in the suburban clinics were significantly more likely to refuse the nasopharyngeal swab than the parents in the inner-city urban clinics. As such, children with missing cultures (who are not included in the study) were more likely to be Caucasian. There were no significant differences in age or gender between children with and without nasopharyngeal cultures. The mean age (standard deviation) of the 204 children included in the present analysis was 6.2 (2.8) years. Younger children were significantly more likely to be colonized with S pneumoniae (38.4% of children
