Guidelines for Acute Sinusitis

Context: Bacterial infections of the paranasal sinuses are an important cause of morbidity in children. However, these infections may be difficult to distinguish from viral upper respiratory infections. Consequently, children with viral upper respiratory infections are frequently treated with antibiotics. The inappropriate use of antibiotics for viral upper respiratory infections is costly and has been responsible, at least in part, for the emergence of antibiotic resistance. To improve the diagnosis and treatment of patients with acute bacterial sinusitis and reduce the inappropriate use of antibiotics, clinical guidelines have been developed. Three national organizations have recently published different guidelines for the diagnosis and treatment of acute bacterial sinusitis: the American Academy of Pediatrics, the Sinus and Allergy Health Partnership and the Centers for Disease Control and Prevention. All three guidelines recommend that the diagnosis and treatment of acute bacterial sinusitis should be based on clinical criteria without the confirmation of imaging or other laboratory data. Although the similarity between the different guidelines suggests that there is widespread consensus to use clinical criteria to diagnose acute bacterial sinusitis, there is virtually no evidence to support this position.

Objectives: The objectives of this study are:

1. to determine the effectiveness of antibiotic treatment of children diagnosed to have acute sinusitis on clinical grounds alone without the performance of sinus images and

2. to evaluate the response to antibiotic therapy or placebo in terms of financial cost to the insurers, patient and family.

Setting and Participants: The study, employing a prospective, randomized, and double blind design will be conducted in three primary care facilities serving a diverse racial and socioeconomic population. Children will be eligible for the study if they are between the ages of 1 and 10 years and present with one of three clinical presentations. Group 1 will be comprised of subjects with persistent disease defined as nasal discharge (of any quality) or daytime cough (which may be worse at night) or both persisting for more than 10 days without evidence of improvement; Group 2 will be patients with non-persistent disease consisting of two subsets: children with nasal discharge or daytime cough that is worsening either on or after the 6th day (> 120 hours) of symptoms [worsening is manifest by the new onset of documented fever (temperature > 100.5o F ) or substantial increase in nasal discharge or cough] and children with severe disease defined as a documented temperature of at least 102o F and purulent (thick, colored and opaque) nasal discharge present concurrently for at least 3 consecutive days (68-72 hours).

Main Outcome Measure: Will be a comparison of the proportion of children who have a complete resolution (cure) of their respiratory symptoms in each of the treatment groups (amoxicillin potassium clavulanate and placebo). Secondary outcome measures will include the proportion of children in each treatment group who:

1. are cured after 72 hours of antibiotic therapy,

2. who fail therapy,

3. who relapse between 10 and 30 days after entry into the treatment groups, and

4. who develop adverse events during the treatment interval of the study. Other outcome measures are the total expenses incurred by the health care system, patients and families during the course of the study and parental impression of whether their child was receiving active drug or placebo.

Statistical Analyses: Analysis will be done, using the Chi square statistic to determine the significance of the difference in proportion of the patients in each treatment group (antibiotic versus placebo) who sustain a complete resolution of their respiratory symptoms. A separate analysis will be done for children with persistent and non-persistent presentations. Analysis will also be done for each of the secondary outcome measures. For children enrolled in the trial, direct medical costs will be calculated and compared between treatment groups, including costs for antibiotic, hospitalizations, emergency room services, non-study related office visits, and antibiotics (other than the study medication), other medications prescribed and over-the counter medications used. Indirect medical costs will be assessed by comparing:

1. number of sick days,

2. number of days in which the parent/s could not work due to the children being sick,

3. number of days in which alternative day care arrangement were necessary for a sick child.

Confounders Included in the Study: The study design, which is prospective, randomized and double blind, will on average, be the most effective way to control for potential confounders. Because the investigators anticipate that age and clinical severity may impact substantially on response to therapy, the groups have been stratified by these two variables.

Biases to be Addressed: The prospective, randomized and double blind study design should minimize the introduction of bias. To insure the generalizability of the results, the study is conducted among three different clinical populations: rural, urban indigent and urban middle class. Strict eligibility criteria and outcome measures will be employed at all three clinical sites in which the study will be performed.