There’s probably no one who has never had a sore throat. And yet, if we’re honest, our approach to this very common complaint in adults—the one we so casually label pharyngitis—is not always as clear as we like to think.

Formally, pharyngitis is a bilateral, diffuse inflammation of the pharynx. In practice, it’s something we see every day—and something we don’t always manage as thoughtfully as we could.

A quick disclaimer: what follows is not a rigid guideline. Local recommendations will vary. I’m also writing with a particular setting in mind—populations at higher risk, where Group A streptococcus is common and access to care for complications may be limited.

But the core question is universal: who, among all these patients with sore throat, actually needs antibiotics?

Let’s first get the obvious—and the dangerous—out of the way. If a sore throat comes with signs of airway compromise, this is no longer a "pharyngitis discussion." This is an emergency. We’re talking about:

• stridor
• trismus
• inability to swallow saliva
• the classic "hot potato" voice
• obvious asymmetry of the neck

You are unlikely to miss these patients. They don’t look well. Fortunately, they are rare—but they matter.

Where it really starts: the history

Time course: how did this start? Quickly, over hours—or slowly over days? Then the epidemiological context:

• recent travel
• vaccination status
• immune competence
• sexual activity/behavior (especially important in atypical presentations)

Then ask a deceptively simple question: what troubles the patient most?

Because the answer often points us in the right direction. Is it the systemic picture—the fatigue, the malaise, that familiar "I’ve been run over by a truck" feeling? Or is it the throat itself?

Are there symptoms involving nearby structures (ears, sinuses, neck)?

A useful reminder: not everything is pharyngitis

It’s easy to anchor too early. But in roughly 10% of cases, what looks like "just a sore throat" turns out to be something else entirely:

• autoimmune diseases (e.g., Crohn’s disease, lupus, Still’s disease)
• drug-related complications like agranulocytosis
• even malignancy, such as leukemic infiltration of the tonsils

There are usually clues—if we look for them:

• mucosal ulcerations extending beyond the throat
• mass lesions

These findings should raise suspicion for diagnoses beyond the more common viral or bacterial pharyngitis.

The uncomfortable truth: viral vs bacterial isn’t always obvious

We like clean distinctions. Reality is less cooperative. Yes, most cases are viral. Yes, bacterial infections matter. But separating the two based on symptoms alone? Not always straightforward—even with experience. Co-infection with both viral and bacterial pathogens is unlikely (although bacterial superinfection following a viral illness is certainly possible).

Several tools are available to assist:

• Centor criteria
• feverPAIN score

However, these tools have important limitations. The Centor criteria, for example, were developed in American adolescents with acute-onset symptoms. They perform poorly when symptoms persist beyond three days. Most importantly, they are not definitive.

And then there’s real life. Viral infections (e.g. COVID-19) may present with anterior cervical lymphadenopathy and, occasionally, even tonsillar exudates. And bacterial infections don’t always read the textbook either

Still, a few patterns tend to hold:

• a thick, dense tonsillar exudate → more consistent with bacterial infection
• cough → more typical of viral illness, reflecting diffuse airway involvement

Do tests solve the problem? Not quite—but they help

If you have access, use them.

• PCR testing for respiratory viruses (COVID-19, influenza A/B, RSV) — results in ~30 minutes
• Rapid antigen detection test (RADT) for Group A β-hemolytic streptococcus — ~30 minutes; RADT is most useful after a negative viral PCR result.
  • high specificity but suboptimal sensitivity (misses 20–30% of Group A cases)
  • does not detect Group C streptococcus or fusobacteria
• Throat culture (if viral tests are negative) — 48–72 hours

Age also matters: children → more commonly Group A streptococcus. In adolescents/young adults Group C may predominate. Both are capable of causing suppurative complications.

A framework useful in practice

It is worth emphasizing that the vast majority of patients have viral pharyngitis.

From a practical standpoint, it is helpful to divide patients into three categories based on how throat pain relates to systemic symptoms.

Category А - The first—and by far the largest. Patients with viral respiratory infections, in whom systemic symptoms and sore throat are roughly equally prominent. This is the familiar "common cold," caused by viruses affecting the entire respiratory tract. In some cases you may consider antiviral drugs (e.g. elderly)

Next are two small groups that require close attention:

Category B — smaller, but clinically important. Patients whose primary complaint is localized throat pain. That’s the main issue. That’s where your suspicion for bacterial infection should rise (often streptococci or fusobacteria).

Category C — also small, but important not to miss. Patients in whom systemic symptoms predominate: poor appetite, sleep disturbance, nausea, rash, lymphadenopathy. Here, pharyngitis is secondary.

This group includes the measles virus and viruses of mononucleosis syndrome (EBV, HIV, CMV). These cases require blood tests, and establishing the diagnosis has important long-term implications.

In rare cases, symptoms of pharyngitis accompany gonorrhea, mycoplasma infection, syphilis, and chlamydia; also opportunistic infections in immunocompromised individuals. In such cases, a careful history is crucial.

Why it matters: complications

Most sore throats will resolve without consequence. But bacterial infections—left untreated—can occasionally lead to complications.

• Non-suppurative complications:
  • rheumatic fever (now rare, but not eliminated)
  • post-streptococcal glomerulonephritis
• Suppurative complications:
  • peritonsillar abscess
  • retropharyngeal abscess (with potential spread to the posterior mediastinum), leading to the development of a severe suppurative process in the chest cavity. Although this complication is more common in children aged 1–6 years, it can also occur in adults.
  • Lemierre’s syndrome (septic thrombophlebitis of neck veins → lungs)
  • Ludwig’s angina (submandibular/sublingual infection)
  • mastoiditis

Sometimes it is not possible to trace the path of bacteria from the site of infection; they may spread into the thoracic cavity as "bacteria escape" resulting in empyema or lung abscesses.

Given the severity of the potential complications, let us ask ourselves the fundamental question:

So—how should we treat?

In practice, treatment decisions are driven primarily by history and clinical presentation, with test results playing a supporting role.

The vast majority of patients with a sore throat will require only symptomatic treatment, as the condition is caused by a viral infection. However, even in clearly viral cases, patients should start feeling better within about 48 hours.

If this does not occur, the diagnosis should be reassessed. A culture should be performed for Group A and non-A (C, G) streptococci, as well as testing for mononucleosis viruses, if this has not been done previously.

And antibiotics?

Antibiotics are justified when: 

1. PCR testing for common respiratory viruses is negative 

2. clinical features strongly suggest bacterial infection. Particularly category B patients - the ones whose main complaint is the throat. 

3. even if RADT is negative, when clinical suspicion remains high

Choice of antibiotics

Fortunately, the usual suspects—Group A and C streptococci, and fusobacteria—are still sensitive to amoxicillin. Duration: 7 days (don’t shorten the course)

If there’s a penicillin allergy: go with clindamycin. Macrolides may seem convenient, but they won’t reliably cover fusobacteria.

Reduce it to one idea

Most sore throats are viral. Some are not. The challenge is not in knowing that—but in recognizing the difference in the patient in front of you

Acknowledgements: Rabih Geha — for creating educational content that inspired this work.