16.4 Common Infections and Anti-Infectives

The current FNP-BC Test Content Outline lists anti-infective agents as an official drug-agent class. Domain III Planning asks whether the guideline supports a drug at all. Domain IV Implementation asks whether the molecule, the dose, the allergy story, and the culture make sense. Respiratory diagnosis already separated bronchitis from pneumonia (Section 11.1) and HEENT already walked AOM observe-versus-treat and sinusitis day-10 rules (Section 15.1). This section is the antibiotic decision that sits on top of those diagnoses.

Stewardship: the infections that do not get a prescription

Most primary-care respiratory illnesses are viral. Do not prescribe antibiotics for:

• Uncomplicated viral upper-respiratory infection (clear or colored mucus, sore throat with cough and rhinorrhea, 3–7 days of cold symptoms)
• Most acute bronchitis in a well adult with a nonfocal exam (Section 11.1)
• Early sinusitis still inside the viral window — colored mucus on day 4 is not bacteria (Section 15.1)

Stewardship is not stinginess. It is how you avoid C. difficile, resistant organisms, rashes you will later call “allergies,” and a population that expects a Z-pak for every winter. Give a specific viral diagnosis, a time course (“cough often lasts two to three weeks”), and the return precautions that make observation safe. A rescue plan for true deterioration is allowed. A 14-day “just in case” amoxicillin course is not.

Syndromes you do treat — and when to culture

Group A strep pharyngitis. Test only when viral features are absent (Section 15.1). Treat a positive RADT with penicillin V or amoxicillin for 10 days (or intramuscular benzathine penicillin G if adherence is doubtful). The point is rheumatic-fever prevention, not a shorter sore throat by 16 hours. Macrolides are for true penicillin allergy, not convenience.

Acute otitis media. Use the AAP age, laterality, and severity table from HEENT. When you treat, high-dose amoxicillin is first-line unless the child had recent amoxicillin or has otitis-conjunctivitis (then amoxicillin-clavulanate).

Acute uncomplicated cystitis in a nonpregnant woman is nitrofurantoin (typically 5 days), trimethoprim-sulfamethoxazole for 3 days when local resistance allows, or single-dose fosfomycin. Do not lead with a fluoroquinolone. Do send a urine culture in pregnancy, in men, in pyelonephritis, after relapse or treatment failure, and in children — those are not “same as last time, no culture” visits. Pyelonephritis (fever, flank pain, nausea) is a different drug and a different disposition if the patient cannot keep fluids down.

Outpatient community-acquired pneumonia. Follow current IDSA adult options and local resistance: in many previously healthy outpatients that is high-dose amoxicillin or doxycycline, not automatic azithromycin (Section 11.2). Hypoxia, hypotension, or a frail older adult who cannot manage oral therapy is an emergency-department pneumonia, not a five-day office experiment.

Cellulitis. Nonpurulent spreading erythema is usually streptococcal — cephalexin or another beta-lactam that covers strep. Purulent cellulitis or abscess is Staphylococcus, including MRSA in many communities: drain the abscess (source control is the treatment), and if you add a drug use trimethoprim-sulfamethoxazole or doxycycline according to local susceptibility. A rapidly spreading toxic infection, crepitus, or orbital involvement leaves the clinic.

Culture when the result changes the drug: recurrent or complicated UTI, pyelonephritis, treatment failure, abscess drainage, gonorrhea/chlamydia NAAT on the right site, and a negative RADT in a child that still needs a backup throat culture. Do not culture every viral pharynx “for completeness.”

Tick disease: the doxycycline teaching that changed

Erythema migrans — the expanding annular rash of early Lyme disease — is a clinical diagnosis in an endemic area. Do not wait for a serology that is still negative in early disease. First-line treatment is doxycycline for a short course (commonly 10 days for uncomplicated EM in current CDC-style adult teaching; follow the labeled duration you are using).

Here is the update older exam banks get wrong: the old absolute ban on doxycycline in children younger than 8 years is outdated for short Lyme courses. AAP and CDC support doxycycline for early Lyme disease in young children because dental staining risk with a short doxycycline course is not the hazard once taught for longer tetracycline exposures. Teach that sentence. Do not swing to the opposite error and use doxycycline like amoxicillin for every pediatric ear infection. For Rocky Mountain spotted fever and other life-threatening rickettsial disease, doxycycline is first-line at any age and delay is deadly.

Amoxicillin remains an alternative for Lyme in pregnancy when doxycycline is avoided. Neurologic Lyme, carditis, and arthritis are not office doxycycline-and-see-you-in-a-month problems.

Antivirals, C. difficile, reportable teasers, and allergy folklore

Influenza antivirals (oseltamivir is the usual primary-care name) help most when started within 48 hours of symptom onset. Treat high-risk outpatients and anyone sick enough for hospital care even if the clock has run past 48 hours. Do not withhold oseltamivir from a day-three pregnant patient because a handout said “only 48 hours.”

COVID-19 outpatient antivirals are for high-risk people and must start early (nirmatrelvir-ritonavir is time-limited, commonly within 5 days of symptom onset, and is full of CYP3A interactions — you will reconcile the med list). Do not invent a unique 2026 product-by-product algorithm if you are not sure of this year’s preferred sequence. Teach eligibility + early start + interaction check + follow current CDC/NIH outpatient criteria.

Clostridioides difficile is the stewardship bill coming due. Watery diarrhea during or after antibiotics — especially clindamycin, fluoroquinolones, and later-generation cephalosporins — needs a stool test, not another cephalosporin. Stop the offending antibiotic when you can. First-line treatment of nonfulminant C. difficile is oral vancomycin or fidaxomicin, not metronidazole as the modern default. Fulminant disease (shock, ileus, megacolon) is an emergency. Proton-pump inhibitors and repeated antibiotics raise risk; so does older age.

Reportable diseases get a teaser here because Implementation will score them again in the regulatory chapter (18.3). In primary care you should already recognize and report (to the local or state health department, not to a blog): measles, tuberculosis, gonorrhea, syphilis, HIV, and pertussis, among others. You do not need the full legal matrix in this section. You do need to know that a morbilliform rash with cough, coryza, and conjunctivitis, a night-sweat upper-lobe cough, or a laboratory-confirmed STI is not a private lifestyle issue you file and forget.

Allergy is how good stewardship dies. Sort the label:

• True IgE: immediate urticaria, angioedema, anaphylaxis, or bronchospasm — take it seriously.
• Delayed maculopapular rash days into a childhood course — often not IgE, often challengeable later.
• Unknown (“my mother said I was allergic”) — not the same as anaphylaxis.

The folklore that 10% of penicillin-allergic patients react to cephalosporins is outdated. Cross-reactivity is much lower, especially with later-generation cephalosporins that do not share an R1 side chain with the culprit penicillin. If the index reaction was anaphylaxis, avoid the identical side-chain pair and use allergy resources or a different class for the first doses. If the index reaction was a mild delayed rash, a dissimilar cephalosporin is often acceptable and is better medicine than a default fluoroquinolone for every cellulitis. Document what happened, when, and which drug. “Allergy: PCN” without a reaction is an incomplete history.

FNP traps: azithromycin for day-4 sinus pressure; treating EM with “watchful waiting for serology”; refusing doxycycline to a 6-year-old with classic Lyme because of a retired tooth-staining rule; starting Paxlovid on day 9 in a low-risk 22-year-old while missing the day-1 high-risk 78-year-old; treating C. difficile with more clindamycin; and converting every unknown penicillin label into lifelong fluoroquinolones because of 1980s cross-reactivity folklore.