Figure 1. Box plot showing a summary of patient age at presentation in pre–pneumococcal conjugate vaccine (PCV7) and post-PCV7 cohorts. The post-PCV7 group was older (P = .007). The horizontal line in the middle of each box indicates the median, and the top and bottom borders of the boxes indicate the 75th and 25th percentiles, respectively. The error bars indicate 95% confidence intervals for each group.
Figure 2. Axial computed tomographic scan of a 14-month-old female with a left medial subperiosteal and orbital abscess (asterisk).
Peña MT, Preciado D, Orestes M, Choi S. Orbital Complications of Acute SinusitisChanges in the Post–Pneumococcal Vaccine Era. JAMA Otolaryngol Head Neck Surg. 2013;139(3):223-227. doi:10.1001/jamaoto.2013.1703
Author Affiliations: Division of Otolaryngology–Head and Neck Surgery (Drs Peña, Preciado, and Choi), Center for Genetic Medicine Research, Children's National Medical Center, Washington, DC (Drs Peña and Preciado); and Department of Otolaryngology–Head and Neck Surgery, Walter Reed National Military Medical Center, Bethesda, Maryland (Dr Orestes).
Importance The widespread use of the 7-valent pneumococcal conjugate vaccine (PVC7), developed to combat invasive Streptococcus pneumoniae infections, has the potential to influence the prevalence and antibiotic resistance patterns of pathogens associated with orbital complications from acute sinusitis. Given the significant morbidity that may result from inadequate treatment of orbital infections related to acute sinusitis, determining the impact of PCV7 on the bacteriology and drug resistance of the pathogens associated with these infections may provide critical information needed to accurately guide optimal clinical management.
Objective To determine if the characteristics of orbital complications from acute sinusitis in children have changed in the post-PCV7 era.
Design Review of clinical data.
Setting Tertiary care children's hospital.
Participants Patients with a diagnosis of orbital cellulitis and/or subperiosteal abscess from January 1, 1996, to December 31, 2009. Patients with immune deficiency or orbital trauma were excluded. Patients were divided into pre-PCV7 (before 2003 [n = 128]) and post-PCV7 (2003 and after [n = 145]) groups. Statistical analyses were used to compare the 2 groups.
Main Outcome Measures Differences in patient demographics, signs and symptoms, laboratory study results, computed tomography scan findings, and microbiological analyses between the pre-PCV7 and post-PCV7 groups.
Results A total of 273 children met the inclusion criteria. The post-PCV7 group was older (71.4 months vs 88.8 months [P = .007]) than the pre-PCV7 group. A significant decrease in S pneumoniae and Streptococcus viridans –positive sinus or blood cultures were observed (22.4% vs 0% [P < .001] and 12.24% vs 0% [P = .005], respectively). An increase in Staphylococcus aureus was seen in the post-PCV7 group (20.4% vs 42.37% [P = .02]). Methicillin-resistant S aureus (MRSA) was isolated only in the post-PCV7 group (P = .002). The pre-PCV7 group had a significantly longer hospital stay than the post-PCV7 group (7.15 days vs 5.47 days [P = .004]).
Conclusions and Relevance Although universal PCV7 vaccination has eliminated S pneumoniae as an etiologic pathogen in acute sinusitis complications in this series, there has been a parallel and significant increase in S aureus, including an increase in the prevalence of MRSA associated with orbital infections related to acute sinusitis.
Despite medical advances in the diagnostic and therapeutic management of orbital complications from acute sinusitis, they continue to be an important clinical problem with the potential to cause considerable morbidity and mortality. The ophthalmologic complications of sinusitis are well documented and include a group of conditions ranging from periorbital inflammation and orbital cellulitis (OC) to subperiosteal and orbital abscess (SPOA/OA) and, finally, cavernous sinus thrombosis.1 With the exception of periorbital inflammation or preseptal cellulitis, patients with these conditions present or develop proptosis, chemosis, ophthalmoplegia, and visual impairment. If not recognized and treated in a timely manner, orbital cellulitis can progress into an orbital abscess, cause blindness, spread into the intracranial cavity, result in systemic sepsis, or even death.
Prior to l985 and the introduction of the Haemophilus influenzae type B vaccine, H influenzae was the most common pathogen associated with clinically significant periorbital infections from acute complicated sinusitis.2 Subsequent studies in the late 1990s documented Streptococcus pneumoniae and other streptococcal species to be the predominant pathogens associated with OC and SPOA/OA and that the incidence of H influenzae had decreased dramatically.3 In an effort to reduce the morbidity of invasive S pneumoniae infections, the US Centers for Disease Control and Prevention recommended the use of the heptavalent pneumococcal conjugate vaccine (Prevnar or PCV7) for children younger than 24 months starting in June 2000.4 A number of publications5,6 have reported a decreased incidence of invasive Spneumoniae infections since then. Most of these references discuss meningitis, and pulmonary and otologic infections, but little is known about the impact on the incidence, presentation, bacterial pathogens, or treatment of orbital complications from acute sinusitis. The purpose of this study was to determine if the characteristics of orbital complications from acute sinusitis have changed in the post–heptavalent pneumococcal conjugate vaccine (PCV7) era.
Institutional review board permission was obtained to conduct this retrospective case series study of patients admitted and treated at a tertiary children's hospital with a diagnosis of orbital cellulitis and/or subperiosteal/orbital abscess (SPOA/OA) from January 1, 1996, to December 31, 2009. Owing to herd immunity, it has been demonstrated that PCV7 has been effective in significantly reducing the invasive diseases caused by S pneumoniae in young children, even if not fully immunized,5- 7 and in adults8 who were not immunized by 2003. Thus, for the purposes of our study, January 1, 2003, was selected as a separation point between the pre-PCV7 and post-PCV7 groups. The post-PCV7 era was defined to begin on January 1, 2003, and to end on December 31, 2009. The pre-PCV7 era was defined to begin on January 1, 1996, and end on December 31, 2002, to allow for equal time for the pre-PCV7 and post-PCV7 groups (7 years each).
To identify all the cases of OC and/or SPOA/OA, the International Classification of Diseases, 10th Revision (ICD-10), code for orbital cellulitis (376.01) was used to search the hospital database for the 14-year study interval. Patients with preseptal cellulitis, orbital trauma, or immune deficiency were excluded. The pre-PCV7 and post-PCV7 groups were compared with respect to their demographic data, presenting signs and symptoms, laboratory study results, computed tomographic (CT) scan findings, and microbiologic analyses. Surgical and medical treatments rendered and length of hospital stay were also compared. Descriptive analyses were conducted to evaluate differences between the groups. We used χ2 or Fisher exact tests to evaluate categorical variables, and t test or Kruskal-Wallis tests were used to evaluate continuous variables. A significance level of .05 was chosen for all analyses.
The diagnosis of OC and/or SPOA/OA was made based on clinical signs, symptoms, and CT scan findings or microbiologic culture results. A total of 430 patients met the initial clinical inclusion criteria based on Current Procedural Terminology code data. However, after complete review of all the medical records available, only 273 were included in the study. These patient medical charts all had CT scan and/or microbiologic culture data supporting the clinical diagnosis of OC and/or SPOA/OA. There were 128 patients in the pre-PCV7 group compared with 145 patients in the post-PCV7 group. Of the 128 patients in the pre-PCV7 group, 49 had either blood or sinus/abscess cultures taken. In the post-PCV7 group, 59 blood or sinus/abscess cultures were collected.
Demographic characteristics and clinical data are summarized in Table 1. Males comprised 62% and 76% of the pre-PCV7 and post-PCV7 groups, respectively. The average age of presentation was 71.4 months in the pre-PCV7 group and 88.8 months in the post-PCV7 group. The post-PCV7 group was older (P = .007), as shown in Figure 1. Pertinent medical historical data and presenting signs and symptoms were similar in the 2 groups. Leukocyte counts at admission were also similar in both cohorts.
All CT scan studies available for review in patients in both the pre-PCV7 and post-PCV7 groups were interpreted by attending radiologists. Opacification of at least 1 paranasal sinus was seen on the CT scan in all patients in the pre-PCV7 and post-PCV7 groups. The proportion of patients with SPOA/OA was similar between the 2 groups. Figure 2 shows a representative CT scan from 1 of these patients. In the pre-PCV7 group, 22 patients had undergone either endoscopic (n = 20) or open drainage procedures (n = 3). One patient had undergone a combination of both endoscopic and open procedures during the same operative setting. Thirty-three patients in the post-PCV7 group underwent 32 endoscopic drainage surgical procedures. Three of these patients had undergone combined open and endoscopic drainage procedures during the same operative setting, and 1 patient had undergone only an open procedure.
Specimens for culture were sent for all surgical patients in both groups (Table 2). No data were available for 1 surgical specimen from the pre-PCV7 group. The technique for collecting the operative cultures was not standardized. Review of the operative reports generally did not discuss the technique for specimen collection or what type of cultures were ordered at the time of surgery. Pathogenic organisms were identified in 41 of 49 cultures taken from the pre-PCV7 group and in 47 of 59 cultures collected from the post-PCV7 group. Streptococcus pneumoniae was the single most common pathogen isolated in the pre-PCV7 group (11 patients [22.4%]; P < .001) but was not isolated in any patients in the post-PCV7 group. Streptococcus viridans was also not detected in any patients in the post-PCV7 group, although it was isolated in 6 patients in the pre-PCV7 group (12.24% vs 0; P = .005). There was no predominant combination of pathogens isolated.
Staphylococcus aureus was more frequently identified in the post-PCV7 group (20.4% vs 42.37%; P = .02). Susceptibility data on S aureus isolated were available in 6 of 10 specimens in the pre-PCV7 group and in all but 1 in the post-PCV7 group (Table 3). Significantly increased methicillin-resistant S aureus (MRSA) was seen in the post-PCV7 group (P < .001). Methicillin-sensitive S aureus was seen in both cohorts.
The PCV7 vaccine was first licensed in the United States in February 2000 for use in children younger than 24 months. The PCV7 vaccine is typically administered at 2, 4, and 6 months with a booster dose given at 12 to 15 months of age and has been shown to significantly decrease the burden of pneumococcal disease in children younger than 5 years. There is also reported evidence of herd immunity in which PCV7 has decreased the levels of Streptococcus pneumoniae below the threshold at which it can be sustained as a pathogen.5 Herd immunity related to the use of PCV7 is due to prevention of nasopharyngeal carriage of vaccine serotype pneumococci. This reduces the transmission of S pneumoniae, which then reduces disease rates even in unvaccinated individuals.9,10 By 2003, approximately 70% of the eligible population had received the 3 doses of the vaccine,11 and herd immunity was evident in adults and unvaccinated children.9,10
More than three-quarters of the patients in this study had documentation that their immunizations were up to date, although PCV7 was rarely specifically mentioned. Since PCV7 has been given to children 24 months or younger starting in 2000, most of the patients in the post-PCV7 group, whose average age was 48 to 50 months, would have received PCV7. Furthermore, as discussed herein, most of the population had either been vaccinated or had developed herd immunity by 2003, so it is likely that most of the post-PCV7 group had immunity against the pneumococcal serotypes contained in PCV7.
The inability to isolate S pneumoniae from either blood or sinonasal abscess cultures post-PCV7 suggests that the vaccine has had a significant impact in reducing invasive S pneumoniae acute complicated sinusitis and associated orbital infections. A similar finding has also been reported in a recent study from the Texas Children's Hospital, where S pneumoniae was found to be the pathogen of interest in 1 of 38 pediatric OC cases.12 Moreover, that study also reported a significant increase in staphylococcal infections, as seen in this investigation. MRSA was identified in 72.3% of the Texas patients. In another study of adult and pediatric sinogenic SPOA/OA, S aureus was the single most commonly identified pathogen with approximately one-quarter of the cases involving MRSA.13 The differences in the rates of MRSA isolated likely reflect the differences in regional distribution of these organisms. The noted predominance of S aureus as a main pathogen in OC/SPOA/OA in the post-PCV7 group, as captured in this analysis, may be potentially explained by a concomitant increase in the virulence of this organism unrelated to the PCV7 vaccine. This virulence has been attributed to the emerging ability of some S aureus strains to produce Panton-Valentine leukocidin, a leukocyte lysing cytotoxin.13
Even though the post-PCV7 group had a greater incidence of more virulent pathogens, as demonstrated by microbiologic analyses, they had shorter hospital stays. The reason for the shorter length of stay in the post-PCV7 group was not readily apparent. It is possible that since younger children tend to respond better to medical treatment,14- 16 and the pre-PCV7 group was younger, that more children in the pre-PCV7 group were initially treated medically compared with the older patients in the latter group. The bacterial culture data, along with the older age at presentation in the post-PCV7 group, may have influenced the decision to operate sooner, thereby decreasing the number of days to surgery. It is also possible that the shorter hospitalization reflects the national trend in decreased length of hospital stay owing to pressures in the current health care system to discharge patients more expediently.17
Streptococcus viridans was only isolated from cultures collected from the pre-PCV7 group. The reason is not entirely clear, as this organism is generally considered a contaminant in immunocompetent patients, and none of the children in this study were immunocompromised. Since S viridans includes many groups of organisms that are either pathogenic or normal flora, it is possible that the S viridans –positive culture results were not sufficiently differentiated to identify pathogens from the Streptococcus milleri group. The latter includes 2 species, Streptococcus intermedius and Streptococcus constellatus, which are commonly associated with abscess formation.18 This seems unlikely because both S intermedius and S constellatus were successfully identified from other pre-PCV7 cultures as well as post-PCV7 cultures. In addition, it is also possible that a better technique for obtaining cultures was practiced in the latter half of the study, although there is no other evidence to support this hypothesis either.
Similar to other retrospective studies, this investigation is limited by selection bias and the quality of the data collected. A number of potential patients had to be excluded from this investigation because the medical records, including the results of imaging studies and microbiologic tests, were not available for review. Furthermore, it is not entirely clear if blood cultures are an ideal surrogate for sinonasal and/or abscess cultures,3,19 even though they may be useful in patients in whom surgical intervention may not be indicated. Nasal cultures could be used instead, but they may be difficult to obtain in uncooperative children and could also be contaminated by staphylococcal colonization of the nasal cavity.13 Finally, the methods used to collect specimens for microbiologic analyses in this study were not standardized. This could have contributed to the confounding results seen with S viridans and had a negative impact on the ability to isolate all pathogenic bacteria, especially with respect to anaerobic bacteria.
In conclusion, although universal PCV7 vaccination has eliminated S pneumoniae as an etiologic pathogen in acute sinusitis complications in this series, there has been a parallel and significant increase in S aureus, including an increase in the prevalence of MRSA associated with orbital infections related to acute sinusitis. Given the significant potential for increased morbidity with OC and SPOA/OA, especially if not treated aggressively in a timely manner, and the increased incidence of MRSA as a potential pathogen for orbital infections from acute complicated sinusitis, it seems prudent to treat patients with these infections with broad spectrum antibiotics that cover MRSA until specific culture information becomes available.
Correspondence: Maria T. Peña, MD, Department of Otolaryngology, Children's National Medical Center, 111 Michigan Ave NW, Washington, DC 20010 (email@example.com).
Submitted for Publication: April 24, 2012; final revision received October 15, 2012; accepted November 26, 2012.
Published Online: February 21, 2013. doi:10.1001/jamaoto.2013.1703
Author Contributions: All authors had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Peña, Preciado, Orestes, and Choi. Acquisition of data: Orestes. Analysis and interpretation of data: Peña, Preciado, Orestes, and Choi. Drafting of the manuscript: Peña. Critical revision of the manuscript for important intellectual content: Preciado, Orestes, and Choi. Statistical analysis: Preciado. Administrative, technical, and material support: Peña and Orestes. Study supervision: Peña and Choi.
Conflict of Interest Disclosures: None reported.
Previous Presentation: This study was presented at the 2012 Annual Meeting for the American Society of Pediatric Otolaryngology; April 22, 2012; San Diego, California.
Additional Contributions: Sherwood Holloway, PA-C, and Michelle Levy, PA-C, provided assistance in data acquisition.