• Understand the indications for antibiotic treatment in acute sinusitis.
• Recognize the potential harms of inappropriate antibiotic prescribing for individual patients and the population at large.
• Review the evidence on the effectiveness of quality improvement efforts to reduce inappropriate antibiotic use.

A healthy 53-year-old woman presented to her primary care physician with upper respiratory symptoms and possible sinusitis. She was prescribed Augmentin (amoxicillin-clavulanate). Despite this therapy, her symptoms persisted. She was then prescribed azithromycin.

Upper respiratory tract infection (URI) symptoms are among the most common presenting complaints to primary care physicians, with 83.1 million visits occurring in 2002 (1) , of which 3.1 million were ultimately ascribed to acute sinusitis in adults.(2) Sinusitis occurs after or in conjunction with a viral URI. Inflammation of the respiratory epithelium lining the paranasal sinuses (most commonly the maxillary sinuses) leads to obstruction of the sinus ostia and accumulation of mucus within the sinuses. The adjacent nasal mucosa is invariably inflamed as well. This process leads to the typical sinus symptoms of headache, nasal congestion and discharge, and facial pain or pressure, sometimes accompanied by sneezing, toothache, or fever.

Most cases of acute sinusitis are caused by viruses, and only 0.5%–2% of cases of viral sinusitis develop into a bacterial infection.(3) However, distinguishing viral from bacterial sinusitis on clinical grounds is difficult, as no single symptom or physical examination finding has been found to be predictive of bacterial sinusitis. The typical symptoms of sinusitis—headache and nasal congestion—do not reliably predict bacterial infection, and imaging studies (such as CT scan or plain radiographs of the sinuses) are frequently abnormal in both viral and bacterial sinusitis. In 2001, the Centers for Disease Control and Prevention (CDC) recommended that acute bacterial rhinosinusitis be diagnosed only when a patient has three clinical criteria (4):

• Maxillary pain or tenderness in the face or teeth.

• Mucopurulent nasal discharge.

• Symptoms have lasted for 7 days or more.

In addition, worsening of symptoms after initial improvement appeared to be moderately predictive of bacterial infection in some studies. A 2007 practice guideline by the American Academy of Otolaryngology—Head and Neck Surgery generally corroborated the CDC guidelines.(5) Both guidelines recommend amoxicillin as the preferred initial antibiotic when antibiotics are warranted, as most cases of bacterial sinusitis are caused by Streptococcus pneumoniae, Haemophilus influenzae, or Moraxella catarrhalis.

Despite these guidelines, overtreatment of acute sinusitis with antibiotics is common. A 2007 study found that antibiotics were prescribed in 82.7% of outpatient visits due to acute sinusitis.(2) Many of these prescriptions are unnecessary, as the vast majority of cases of sinusitis are viral in origin—especially when symptoms have lasted for less than 1 week.

In this case, the primary care physician should have asked the patient about the duration of symptoms, character of nasal discharge, and presence of toothache, and examined her for evidence of tenderness over the maxillary sinuses. Antibiotic treatment with amoxicillin would have been justified if the three clinical criteria above were present. If antibiotics were not warranted, management should have focused on symptomatic therapy, including decongestants and antiinflammatory agents.

The patient was prescribed Augmentin (amoxicillin-clavulanate) as initial therapy. While this agent is the second most common antibiotic prescribed for acute sinusitis (behind amoxicillin) (2) , its choice in this scenario illustrates another facet of inappropriate antibiotic use: prescribing of broad-spectrum agents when narrow-spectrum antibiotics are indicated. The use of broad-spectrum antibiotics rose significantly during the 1990s. For sinusitis, prescribing of broad-spectrum agents increased from less than 20% (of cases where antibiotics were prescribed) in 1991 to more than 40% in 1999.(6) Both amoxicillin-clavulanate and azithromycin are considered broad-spectrum antibiotics, and neither has been demonstrated to be significantly more effective at curing sinusitis compared with amoxicillin. Even if antibiotics had been warranted in this case—which is unlikely—treatment should have consisted of amoxicillin along with symptomatic therapies. A second antibiotic course could be justified only if infection with a resistant organism was suspected, which would be unlikely in a previously healthy patient with no recent history of antibiotic use.

Shortly after starting her second course of antibiotics, the patient began feeling unwell. A few days later, she was found down in her home by her daughter. The patient was brought to the emergency department for evaluation. Her work up revealed profound anemia due to brisk autoimmune hemolysis. This was thought to be due to the amoxicillin-clavulanate she had received. She was started on high-dose immunosuppressive therapy with steroids.

Although antibiotics have yielded undeniable benefits for patients since their introduction into medical practice, inappropriate use of these agents results in adverse effects for both individuals and the population at large. Beta-lactam antibiotics such as amoxicillin are generally quite safe, but prescribers and patients must be aware of a wide range of potential adverse effects, ranging from common problems like antibiotic-associated diarrhea (which can occur in up to 34% of patients receiving a typical course of amoxicillin-clavulanate), to rare but dangerous reactions such as Clostridium difficile colitis, anaphylaxis, or this patient's problem: autoimmune hemolysis. Many antibiotics may cause drug-induced autoimmune hemolytic anemia; in the case of penicillins, the mechanism is generally via formation of drug-specific IgG antibodies in the patient's serum, resulting in a direct antiglobulin (Coombs') positive hemolytic anemia.(7) Amoxicillin was first recognized as a cause of autoimmune hemolytic anemia more than 2 decades ago.(8) Although mild cases may be managed by withdrawal of the antibiotic, cases of severe symptomatic anemia require treatment with high-dose glucocorticoids, as in this patient.

The chief population-level effect of antibiotic overuse is the widespread and growing problem of antimicrobial resistance (AMR). AMR is a worsening problem among many bacteria, including Staphylococcus aureus, Streptococcus pneumoniae, and Escherichia coli—organisms that cause common clinical syndromes such as cellulitis, community-acquired pneumonia, and urinary tract infection. Once confined to hospitals, these drug-resistant pathogens are becoming increasingly prevalent in the community setting, and some data indicate that prior treatment with antibiotics may increase an individual patient's likelihood of contracting an infection with a drug-resistant bacteria.(9) AMR exerts significant societal costs, as infections with drug-resistant bacteria are associated with increased morbidity, mortality, and health care expenditures.

Antibiotic use leads to AMR by two mechanisms: creation of a susceptible host by eliminating an individual's normal bacterial flora and selective pressure promoting survival of bacterial strains with genetic mutations that confer antibiotic resistance.(10) Due to this close link between antibiotic prescribing and development of AMR, extensive national and international efforts (11) have focused on reducing antibiotic prescribing for conditions in which antibiotics are not usually indicated. The CDC's "Get Smart" campaign is a prominent example.(12) A major focus of these efforts is reducing antibiotic prescribing for acute respiratory infections (ARIs), including sinusitis, as these infections are rarely bacterial in origin.

The patient's hospital course was marked by multiorgan failure, septic shock, and spontaneous bowel perforation requiring hemicolectomy. Examination of the bowel showed Aspergillus, leading to a diagnosis of disseminated aspergillosis. Despite aggressive antifungal therapy, the patient ultimately succumbed to overwhelming infection and died.

This patient suffered a tragic outcome likely related to inappropriate prescribing of antibiotics. While the complications and ultimate outcome of this case are exceedingly rare, unfortunately, the problem of inappropriate antibiotic prescribing remains common. Over the past decade, antibiotic prescribing for ARIs has decreased in response to publicity and education regarding antimicrobial resistance. However, prescribing rates for viral infections remain high: in 2002, nearly half of adults with nonspecific ARIs were still prescribed antibiotics.(13) Limited success in reducing overall antibiotic prescribing may be counteracted by a marked increase in prescribing of broad-spectrum antibiotics, the use of which doubled during the 1990s.(6)

A clinician's decision to prescribe antibiotics is the result of several factors, including patient factors (patients often expect to be prescribed antibiotics to treat respiratory infections), physician factors (physicians often use heuristics to judge if antibiotics are warranted, rather than relying on evidence-based criteria), and health care system factors (requiring prior approval for acute appointments may result in fewer visits for respiratory symptoms, and correspondingly fewer antibiotic prescriptions).(10) Quality improvement (QI) efforts to reduce inappropriate antibiotic prescribing have used various methods to educate patients and clinicians on indications for antibiotic prescribing. Providing targeted feedback to clinicians on their prescribing practices has also been used. Community-wide campaigns, using mass media communications and other strategies to simultaneously target patients and clinicians, are underway in several European countries and US states.

Reviews of published QI efforts show them to be moderately effective in reducing inappropriate prescribing and reducing unnecessary broad-spectrum antibiotic use.(10, 14) While no single strategy appears uniquely effective, promising strategies include mass media campaigns in combination with targeted clinician education and use of explicit clinical decision support algorithms to indicate when antibiotic prescribing is appropriate. A decision support system could have been very useful in this case. In such a system, the clinician would have been prompted to enter the patient's presenting symptoms and signs, and the system would provide patient-specific treatment recommendations. A recent cluster-randomized trial using a handheld computer–based decision support system for prescribing in respiratory infections accomplished significant community-wide reductions in antibiotic use in communities in Utah and Idaho.(1)

Antibiotic prescribing for respiratory symptoms is frequently driven by a physician's desire to respond to a patient's explicit (or implied) request for antibiotics.(16) However, research has shown that even patients who explicitly request antibiotics are satisfied if clinicians directly address their concerns by explaining the rationale for not prescribing antibiotics and offer symptomatic therapy instead.(17) QI efforts to reduce antibiotic prescribing have not caused increased dissatisfaction with care.(10)

The tragic clinical outcome of this case is undoubtedly rare, but if inappropriate antibiotic prescribing continues unchecked, the societal costs may be equally dramatic. The marked rise in infections caused by methicillin-resistant Staphylococcus aureus (MRSA) (18) is but one example of the clinical implications of drug-resistant bacteria—a problem that will undoubtedly worsen if indiscriminate antibiotic use continues. Despite some successes, inappropriate antibiotic prescribing remains widespread, and clinicians must take responsibility for improving their prescribing practices. Although on the surface this case may appear to be an example of "cascade iatrogenesis" (19) rather than a true medical error, failure to adhere to evidence-based treatment guidelines is increasingly being treated as an error.(20) The burden of responsibility lies on clinicians to practice judicious antibiotic prescribing in order to avoid considerable health implications for their patients in the future.

Take-Home Points

• Inappropriate antibiotic prescribing remains common, especially for acute respiratory infections.
• Clinicians should follow evidence-based treatment guidelines for sinusitis.
• Community-wide campaigns and clinician decision support systems show promise as means of addressing the overprescribing of antibiotics.

Sumant Ranji, MD
Assistant Clinical Professor, Division of Hospital Medicine
University of California, San Francisco

Faculty Disclosure: Dr. Ranji has declared that neither he, nor any immediate member of his family, has a financial arrangement or other relationship with the manufacturers of any commercial products discussed in this continuing medical education activity. In addition, his commentary does not include information regarding investigational or off-label use of pharmaceutical products or medical devices.

1. Woodwell DA, Cherry DK. National Ambulatory Medical Care Survey: 2002 summary. Adv Data. August 2004:1-44. [go to PubMed]

2. Sharp HJ, Denman D, Puumala S, Leopold DA. Treatment of acute and chronic rhinosinusitis in the United States, 1999-2002. Arch Otolaryngol Head Neck Surg. 2007;133:260-265. [go to PubMed]

3. Piccirillo JF. Acute bacterial sinusitis. N Engl J Med. 2004;351:902-910. [go to PubMed]

4. Hickner JM, Bartlett JG, Besser RE, et al. Principles of appropriate antibiotic use for acute rhinosinusitis in adults: background. Ann Intern Med. 2001;134:498-505. [go to PubMed]

5. Rosenfeld RM, Andes D, Bhattacharyya N, et al. Clinical practice guideline: adult sinusitis. Otolaryngol Head Neck Surg. 2007;137(suppl 3):S1-S31. [go to PubMed]

6. Steinman MA, Gonzales R, Linder JA, Landefeld CS. Changing use of antibiotics in community-based outpatient practice, 1991-1999. Ann Intern Med. 2003;138:525-533. [go to PubMed]

7. Gehrs BC, Friedberg RC. Autoimmune hemolytic anemia. Am J Hematol. 2002;69:258-271. [go to PubMed]

8. Gmür J, Wälti M, Neftel KA. Amoxicillin-induced immune hemolysis. Acta Haematol. 1985;74:230-233. [go to PubMed]

9. Moellering RC Jr, Graybill JR, McGowan JE Jr, Corey L, and American Society for Microbiology. Antimicrobial resistance prevention initiative--an update: proceedings of an expert panel on resistance. Am J Infect Control. 2007;35:S1-S23; quiz S4-S6. [go to PubMed]

10. Ranji SR, Steinman MA, Shojania KG, et al. Antibiotic Prescribing Behavior; vol 4. In: Shojania KG, McDonald KM, Wachter RM, Owens DK, eds. Closing the Quality Gap: A Critical Analysis of Quality Improvement Strategies. Technical Review 9. Rockville, MD: Agency for Healthcare Research and Quality; 2006. AHRQ Publication No. 04-0051-4.

11. Goossens H, Guillemot D, Ferech M, et al. National campaigns to improve antibiotic use. Eur J Clin Pharmacol. 2006;62:373-379. [go to PubMed]

12. Interagency Task Force on Antimicrobial Resistance. A Public Health Action Plan to Combat Antimicrobial Resistance. Centers for Disease Control and Prevention Web site. Available at: http://www.cdc.gov/drugresistance/actionplan/html/index.htm.

13. Roumie CL, Halasa NB, Grijalva CG, et al. Trends in antibiotic prescribing for adults in the United States--1995 to 2002. J Gen Intern Med. 2005;20:697-702. [go to PubMed]

14. Steinman MA, Ranji SR, Shojania KG, Gonzales R. Improving antibiotic selection: a systematic review and quantitative analysis of quality improvement strategies. Med Care. 2006;44:617-628. [go to PubMed]

15. Samore MH, Bateman K, Alder SC, et al. Clinical decision support and appropriateness of antimicrobial prescribing: a randomized trial. JAMA. 2005;294:2305-2314. [go to PubMed]

16. Wilson AA, Crane LA, Barrett PH, Gonzales R. Public beliefs and use of antibiotics for acute respiratory illness. J Gen Intern Med. 1999;14:658-662. [go to PubMed]

17. Mangione-Smith R, McGlynn EA, Elliott MN, Krogstad P, Brook RH. The relationship between perceived parental expectations and pediatrician antimicrobial prescribing behavior. Pediatrics. 1999;103:711-718. [go to PubMed]

18. Klevens RM, Morrison MA, Nadle J, et al. Invasive methicillin-resistant Staphylococcus aureus infections in the United States. JAMA. 2007;298:1763-1771. [go to PubMed]

19. Hofer TP, Hayward RA. Are bad outcomes from questionable clinical decisions preventable medical errors? A case of cascade iatrogenesis. Ann Intern Med. 2002;137:327-333. [go to PubMed]

20. Wachter RM, Foster NE, Dudley RA. Medicare's decision to withhold payment for hospital errors: the devil is in the details. Jt Comm J Qual Patient Saf. 2008;34:116-123. [go to PubMed]