Healthcare-Associated Sinusitis
Marc J.M. Bonten
Healthcare-associated sinusitis (HAS) is a common, unrecognized cause of fever and even sepsis in mechanically ventilated patients. Underestimation of its incidence is at least partly due to the difficulty in diagnosing HAS. The reported cumulative incidence ranges from 1% to 83% in studies specifically designed to investigate HAS. Combined with pneumonia, catheter-related sepsis, and urinary tract infection, HAS has been considered as one of the four “horsemen” of clinically important healthcare-associated infections in critically ill patients (1). HAS is most often caused by enteric gram-negative bacteria or Staphylococcus aureus. The infection is a result of disturbances of local anatomy, colonization of the upper respiratory tract with potentially pathogenic microorganisms, and the severity of underlying illness in critically ill patients. The most important risk factors are prolonged nasotracheal intubation, mechanical ventilation, and the presence of a nasogastric tube. Basic infection control procedures and avoidance of nasotracheal intubation seem to be most important for prevention of HAS.
DEFINITION
Because of the wide variation of definitions used for HAS, interpretation of the whole body of literature dedicated to this topic is difficult, mainly because of the problems encountered in diagnosing HAS. It is usually diagnosed using a combination of clinical suspicion of infection, with fever and leukocytosis, together with radiologic evidence of HAS. The latter may be based on radiographic, ultrasonographic, or computed tomography (CT) examinations. Finally, the diagnosis is confirmed by microbiologic cultures. The value of each of these diagnostic modalities is discussed later. In what probably is the most detailed prospective study of HAS, Rouby et al. (2) distinguished between radiologic maxillary sinusitis and infectious maxillary sinusitis (Fig. 23-1). Radiologic maxillary sinusitis was defined as total opacification of one or both maxillary sinuses or as the presence of an air-fluid level within one or both maxillary sinuses on CT image. Based on microbiologic cultures and Gram staining, the diagnosis of infectious maxillary sinusitis was established or refuted.
CLINICAL RELEVANCE OF HEALTHCARE-ASSOCIATED SINUSITIS
HAS was first described in 1974. Arens et al. (3) described four patients who had undergone nasotracheal intubation for coronary artery bypass surgery and who developed HAS. All patients had been intubated <36 hours, and evidence of HAS appeared in 6 to 10 days postoperatively. In later studies, HAS was usually described in patients who were still intubated. The true incidence of HAS and its relevance as a source of fever are unknown. Large studies determining prevalences and incidences of healthcareassociated infections, such as the National Nosocomial Infection Surveillance system from the Centers for Disease Control and Prevention or the European Prevalence of Infection in Intensive Care Study, did not include HAS as an infectious entity (4,5). However, the cumulative incidence of HAS was remarkably high in several studies carefully analyzing causes of fever in mechanically ventilated patients, with the reported cumulative incidence ranging from 1% to 83% (Table 23-1). Meduri et al. (16) subjected 50 patients with a clinical suspicion of ventilator-associated pneumonia to a systematic diagnostic protocol, which included CT scanning of the sinuses and aspiration of the maxillary sinuses for microbiologic analysis when air-fluid levels or opacifications were encountered. A definitive source of fever was identified in 45 patients and HAS was diagnosed in 12 of them. HAS was in all cases accompanied by another infection, which in most cases (72%) was caused by pathogens other than those isolated from maxillary aspirates. In a large prospective study in medical intensive care unit (ICU) patients with endotracheal intubation, the cumulative incidence of HAS was 7.7%, with incidence rates of 12 cases per 1,000 patient-days and 19.8 cases per 1,000 nasoenteric tube days (11). Furthermore, cumulative incidences from 9% to 26% have been reported in neurosurgical ICU patients (13,14,18).
These studies suggest that HAS may occur often in selected patient groups. However, it is unknown to what extent HAS affects morbidity and patient outcome. Interestingly, HAS may occur concomitantly with other infections. For instance, Borman et al. (21) described 19 patients with
radiographic evidence of HAS, and 10 of these patients had positive cultures of antral aspirates. Evaluation of causes of fever in these patients revealed that fever was definitely caused by HAS in only one patient, possibly in two, and definitely not caused by HAS in the remaining 16 patients.
radiographic evidence of HAS, and 10 of these patients had positive cultures of antral aspirates. Evaluation of causes of fever in these patients revealed that fever was definitely caused by HAS in only one patient, possibly in two, and definitely not caused by HAS in the remaining 16 patients.
 FIGURE 23-1 Possible diagnostic track for patients with a clinical suspicion of HAS. (Modified from Rouby JJ, Laurent P, Gosnach M, et al. Risk factors and clinical relevance of Healthcare-associated maxillary sinusitis in the critically ill. Am J Respir Crit Care Med 1994;150:776-783, with permission). |
CLINICAL MANIFESTATIONS AND DIAGNOSIS
Clinical Presentation
In previously healthy and ambulatory patients, acute sinusitis usually results in localized pain, nasal congestion, and purulent nasal drainage. Sinus disease is an inherent part of the common cold syndrome, and 87% of ambulatory patients with colds have sinus cavity disease (22). In these patients, sinusitis is rarely associated with systemic symptoms or fever (23). Pain cannot be expressed by most intubated patients, and findings of physical examination, such as tenderness and purulent nasal discharge, are often absent. As a result, physical examination usually does not contribute to establishing the diagnosis of HAS (12). Nonspecific symptoms such as fever or leukocytosis often are the first signs of HAS. Because fever and leukocytosis, in this patient population, may have many other causes, both infectious and noninfectious, HAS may not be considered as the cause of infection by clinicians. Careful radiographic and microbiologic analyses are, therefore, mandatory.
Radiologic Examination
Sinus radiography usually includes three views (24): the straight anterior-posterior view (Caldwell view) for examining the frontal and ethmoid sinuses; the Water’s view to visualize the maxillary sinuses (also a straight anteriorposterior view with the patient’s head tilted upward); and the lateral view to visualize the sphenoid sinus. Because of the complex labyrinthine structure of air cells separated by bony septa, the ethmoid sinus is difficult to evaluate. In addition, the sphenoid sinus is localized centrally and surrounded by bony structures and, therefore, is also difficult to evaluate. In critically ill patients, the diagnostic yields of conventional radiography are further diminished by the use of portable equipment, difficulties in placing patients in the upright position, and interference of nasogastric and nasotracheal tubes with x-ray images. Conventional multiview plain sinus radiographs, therefore, are regarded as inaccurate for diagnosing HAS (6).
Computed axial tomography displays bony details and can distinguish soft tissue swelling or fluid within the sinuses. In healthy subjects, sinuses are aerated. Signs suggestive for infection include maxillary mucosal thickening, total opacification, or the presence of an air-fluid level in one or both maxillary sinuses (2). CT scanning definitely has multiple advantages over conventional radiography for diagnosing HAS. However, mucosal thickening or fluid accumulation within sinus cavities are not proof of infection, and CT scan is unable to distinguish between blood and other fluids, which may be problematic in patients
with facial trauma. Even total opacification of one or both maxillary sinuses or an air-fluid level within one or both maxillary sinuses had specificities for infectious maxillary sinusitis ranging from 38% to 69% (2,15,21). Furthermore, CT scan is costly and requires transport of patients, which may, in itself, be a risk factor for healthcare-associated infections (25).
with facial trauma. Even total opacification of one or both maxillary sinuses or an air-fluid level within one or both maxillary sinuses had specificities for infectious maxillary sinusitis ranging from 38% to 69% (2,15,21). Furthermore, CT scan is costly and requires transport of patients, which may, in itself, be a risk factor for healthcare-associated infections (25).
TABLE 23-1 Cumulative Incidence of Healthcare-Associated Sinusitis According to Patient Population and Diagnostic Techniques Used | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Bedside sinus ultrasonography may be a reliable, noninvasive, and cheap alternative to CT scanning. This method, when compared with culture of antral aspirates as a gold standard, has been demonstrated to be accurate in ambulatory adults and children (26). However, clinical experience in mechanically ventilated patients is limited (6,14,27,28). In one study, 100 patients were examined with bedside sinus ultrasonography on admission and every 48 hours thereafter. CT scanning of the head was performed at the discretion of attending physicians and was performed in 61 patients. Fifteen patients had fluid within the maxillary sinus detected by ultrasonography, and in nine other patients sinus fluid was detected by a head CT scan but not by bedside sinus ultrasonography. None of these nine patients, however, had clinical sepsis without another clearly documented source. The authors concluded that the head CT scan is more sensitive but may detect abnormalities that have little clinical significance (6). In another study, left and right paranasal sinuses were examined by ultrasonography in the supine and semirecumbent position in 15 neurosurgical ICU patients in whom HAS was suspected on clinical grounds. Findings of ultrasonography were compared with observations made by sinoscopy. Sensitivities of ultrasonography for the presence of fluid and edema were higher in the semirecumbent position (91% and 81%, respectively). However, specificity was only 25% for the presence of fluid. Moreover, edema and/or secretions were demonstrated in 29 of 30 sinus cavities examined, but microorganisms were cultured from only two antra (14). In a third study, A-mode ultrasonography of maxillary and frontal sinuses was performed in 50 comatose patients that needed cerebral CT for another reason than suspicion of sinusitis (28). With CT images as gold standard, ultrasonography had a specificity of 72% to 98% and sensitivity of 63% to 86% for maxillary sinuses, and of 96% to 99% and 14% to 57%, respectively, for frontal sinuses. With areas under the receiver-operating characteristic curves of 0.89 and 0.76, for maxillary and frontal sinuses, respectively, the authors concluded that ultrasonography was an accurate tool to detect secretions in maxillary sinuses (28). In addition, excellent agreement levels (with kappa statistic >0.9) between B-mode ultrasonographic examination of both maxillary sinuses and CT imaging have been reported (27). In a subsequent study, it was demonstrated that ultrasound evidence of sinusitis was highly predictive for receiving fluids (for microbiological cultures) after transnasal puncture (29). These data suggest that ultrasonography may be a useful screening test, but whether it can be used as the sole diagnostic method remains to be established.
Microbiologic Analysis
The problem of microbiologic analyses in many ICU-acquired infections is distinguishing between colonization and infection. Colonization of the upper respiratory tract (e.g., nares, oropharynx, and trachea) is universal in mechanically ventilated patients. Nasal swab cultures will grow upper respiratory tract flora and are believed to be of little value to determine pathogens causing sinusitis (30). Mucociliary clearance and drainage may keep the sinuses clean. Therefore, antral aspirate cultures are regarded as the gold standard. The frontal, ethmoid, and sphenoid sinuses can only be drained surgically and are not amenable to aspiration at the bedside. However, the maxillary sinuses can be drained, and these cavities are most often involved. CT imaging demonstrated that the maxillary sinuses are involved in almost all ICU patients who develop HAS, and radiographic evidence of maxillary sinusitis was associated with radiologic abnormalities of ethmoid and sphenoid sinuses in >80% of ventilated patients. However, according to Rouby et al.’s (2) study, 50% of the patients with normal maxillary sinuses on CT had radiologic signs of ethmoid and/or sphenoid sinusitis, as did 92% of patients with mucosal thickening in maxillary sinuses. The contribution of infection of the ethmoid and sphenoid sinuses has never been studied.
Aspiration cultures from maxillary sinuses are representative for microorganisms causing pansinusitis, and irrigation at this site is often therapeutic. Insertion is performed with a specialized trocar, which has an inner needle obturator with an outer sleeve. Once inserted, the needle can be removed and irrigation can be performed via the hollow sleeve (24). Because of colonization of the nares, even transnasal cultures can be falsely positive because of introduction of pathogens into the sinus cavity. Adequate disinfection, therefore, has been recommended (2). Disinfection of the nares with a povidone-iodine solution proved to be totally adequate (sterile cultures) in 51%, partially effective (decrease in nasal bacterial burden) in 38%, and completely ineffective (increase in nasal bacterial burden) in 11% (2). In Rouby et al.’s study, patients underwent transnasal puncture of the affected maxillary sinus after nasal disinfection. The diagnosis was changed to infectious maxillary sinusitis when there were more than five polymorphonuclear leukocytes per oil immersion field and a positive culture from sinus aspirate. In patients who did not receive antibiotics, the diagnosis of infectious maxillary sinusitis was established by quantitative cultures depending on the effectiveness of nasal disinfection (cutoff points were >103 colony forming units [CFU]/mL with adequate nasal disinfection [sterile nasal swab] and >104 CFU/mL with inadequate nasal disinfection [positive nasal swab]) (2). Two studies reported poor correlations between endoscopically guided middle meatal cultures and cultures from antral lavage aspirates or taps in patients with clinical suspicion of HAS (31,32).
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
