Community-acquired pneumonia — the Pakistan pathway
The national guideline was rewritten in March 2026 and the outpatient macrolide is gone. Here is the current algorithm, the resistance data behind it, and the parts it still leaves to your judgement.
The macrolide came out of first line in March 2026
The previous Pakistan Chest Society CAP guideline offered "amoxicillin or a macrolide" to a previously healthy outpatient, and doctors reading it against Pakistani pneumococcal data had to work out for themselves that only one of those two options still held. The March 2026 edition makes the call itself: the outpatient first line is amoxicillin or doxycycline, and macrolide monotherapy is explicitly ruled out.
Avoid macrolide monotherapy due to high resistance (macrolide resistance >30% in S. pneumoniae).
Azithromycin has not disappeared from the guideline — it stays, as the second agent alongside a beta-lactam in inpatients. What it is no longer is a drug you send an outpatient home on by itself.
Every recommendation here carries a divergence badge.Same as international guidelinePakistan-adaptedLocal data onlyNational guideline outdated
Most of this page is badged Pakistan-specific rather than adapted, because the March 2026 guideline is built on Pakistani surveillance rather than lifted from international guidance — its own words are that regimens designed for high-income countries "are often inappropriate" here.
Decision-support only. If your hospital produces an antibiogram, use it and ignore this page for empiric choice. If it does not, this is the argument for producing one — and the two series here differ enough that regional data genuinely changes the answer.
✓ Clinically reviewed by Dr Mehdi Raza, MBBS · 06 Aug 2026.
Step 1 · Is it pneumonia — and is it bacterial?
Pakistan Chest Society, March 2026
Imaging is part of the definition, not an optional extra Same as international guideline
In addition to suggestive clinical features, a demonstrable infiltrate by chest radiograph or other imaging technique, with or without supporting microbiological data, is required for the diagnosis of pneumonia.
A febrile cough with crackles and a clear film is not pneumonia, and treating it as such is a large share of the avoidable antibiotic use driving the resistance figures in step 4.
The differential that matters most here Pakistan-adapted
Pakistan carries one of the world's highest tuberculosis burdens. Subacute onset, weight loss, night sweats, haemoptysis, upper-lobe or cavitating changes, or failure to respond to an appropriate antibiotic should all prompt sputum testing for TB rather than a second course of antibiotics. Fluoroquinolones are a particular trap: they have antimycobacterial activity, so an empiric respiratory fluoroquinolone can partially treat undiagnosed TB, blunt the clinical picture, delay diagnosis and select for fluoroquinolone-resistant TB.
Why it is flagged: International CAP guidelines are written for settings where TB is rare. Here it is not, and the fluoroquinolone interaction with TB is a genuine reason to prefer a beta-lactam backbone. The 2026 guideline names overuse of fluoroquinolones as one of its four markers of inappropriate empiric therapy.
What CAP does in Pakistan Local data only
75–120 per 100,000 adults per year
Incidence — Extrapolated from tertiary-hospital studies and provincial surveillance; the guideline says the true incidence is underestimated. Seasonal peaks November to February.
9.5%
In-hospital mortality — Multi-centre cohort 2022–2024 across five provinces.
Sindh 11.2% vs Punjab 7.8%
Provincial spread — Same cohort — a 3.4-point gap in death rate between provinces.
40–45% of hospitalised patients
Drug-resistant organisms — ESBL-producing Klebsiella pneumoniae or Pseudomonas aeruginosa, on Pakistan-specific surveillance.
Factors the guideline links to higher mortality
Age over 65
Chronic lung disease — COPD, bronchiectasis
Diabetes mellitus, chronic kidney disease
Indoor air pollution — solid biomass use
Tobacco exposure — national smoking prevalence around 27%
Multi-lobar radiological involvement
Male sex — male:female ratio around 1.4:1, attributed to differential smoking and occupational exposure
Step 2 · Severity and site of care
Pakistan Chest Society, March 2026 — chapter 4
CURB-65 to decide site of care
The CURB-65 score is widely recommended by international and regional guidelines, including those from the BTS, Indian Chest Society, and PCS Pakistan.
Confusion
Urea above 7 mmol/L
Respiratory rate 30/min or more
Blood pressure — systolic below 90 mmHg or diastolic 60 mmHg or below
Age 65 years or over
0–1
Low risk — outpatient management
2
Consider short hospital stay or close monitoring
3–5
High risk — hospitalisation or ICU
The score is not the whole decision. A nationwide study found disagreement between the CURB-65 recommendation and the physician's actual admission decision in almost four out of every ten patients. The score informs the decision; it does not make it — and in a setting where a patient may not be able to return, that judgement carries real weight.
CRB-65 where there is no lab Pakistan-adapted
The CRB-65 version omits serum urea, which makes it feasible without immediate lab access. The guideline names this as particularly useful in rural Pakistani setups and Basic Health Units — a scoring tool you can complete at the bedside with a watch and a blood-pressure cuff.
Where the score under-calls it — the guideline's own local caveats Local data only
Young patients with multi-lobar disease may be under-triaged by CURB-65.
Malnourished or diabetic patients may have poorer outcomes despite low scores.
CRP and procalcitonin are rarely available at point of care in Pakistan, so biomarker-assisted triage is mostly theoretical here.
Escalate regardless of score. Oxygen saturation below 90%, confusion, or a respiratory rate of 30 or more should trigger immediate escalation regardless of score. Some hospitals in Karachi now pair CRB-65 with SpO₂ below 90% as a single red flag for immediate referral.
The Pneumonia Severity Index is more comprehensive but has limited utility in most Pakistani hospitals — complexity, the need for laboratory and radiographic data, and time burden in a busy emergency room. SMART-COP and SCAP have no large-scale Pakistani validation. CURB-65 and CRB-65 remain the preferred tools.
Which category is this patient?
Step 3 · The national algorithm
Pakistan Chest Society — Community Acquired Bacterial Pneumonia, March 2026
Local data only Current national guidance, and the one to work from. Unlike the edition it replaces, it is built on Pakistani surveillance rather than adapted wholesale from high-income guidance — its own words: antibiotic regimens designed for high-income countries "are often inappropriate due to Pakistan's high rates of antimicrobial resistance and significant burden of Gram-negative organisms".
Drug-resistant pathogen risk is the branch point Local data only
Pakistan-specific surveillance reveals that 40–45% of hospitalized CAP patients harbor drug-resistant organisms, such as ESBL-producing Klebsiella pneumoniae or Pseudomonas aeruginosa.
Hospitalisation within the past 90 days
Prior antibiotic use in the past 3 months
Chronic lung disease — COPD, bronchiectasis
Structural lung disease, chronic kidney disease, or malignancy
Tube feeding or aspiration risk
Known colonisation with multidrug-resistant organisms
Two or more of these risk factors call for extended-spectrum or dual therapy.
Why it is flagged: International CAP guidance treats drug-resistant Gram-negatives as a hospital-acquired problem. Pakistani surveillance puts them in nearly half of admitted community cases — the guideline attributes part of that to healthcare-associated strains being misclassified as community-acquired, which is exactly why it asks for a deliberate risk assessment rather than a default regimen.
Outpatients — 5 to 7 days if stable and clinically improving
Inpatients — 7 to 10 days depending on severity
Five days or fewer may be used in mild to moderate disease if procalcitonin-guided and the patient is recovering rapidly — which in most Pakistani settings means it is not an option, because procalcitonin is rarely available
The guideline's own audit found duration exceeded recommended limits in 42% of cases. Stopping on time is one of the few stewardship levers that costs nothing.
Cautions the guideline attaches
Fluoroquinolones
Avoid in patients at risk of QT prolongation or with a cardiovascular history. Separately, and specific to Pakistan, they partially treat undiagnosed tuberculosis — see step 1.
MRSA suspected
Add vancomycin 15 mg/kg IV twice daily, or linezolid 600 mg IV twice daily.
Pseudomonas suspected
Avoid ceftriaxone; use an anti-pseudomonal beta-lactam.
Step 4 · The resistance data underneath it
Country data on AMR in Pakistan in the context of community-acquired respiratory tract infections. Susceptibility figures are 2015–17, CLSI breakpoints.
What the 2026 guideline itself reports. Macrolide resistance in S. pneumoniae "exceeds 25%, which is the threshold used in international guidelines to discourage monotherapy" — and it notes that macrolides "remain commonly prescribed empirically, particularly in outpatient settings". For H. influenzae: ampicillin resistance 18–25%, macrolide resistance around 15–20%, and co-amoxiclav and cefuroxime above 95% susceptibility.
Streptococcus pneumoniaeLocal data only
The commonest cause of adult CAP in Pakistan. Susceptibility 2015–17, CLSI breakpoints; penicillin and cefaclor both showed declining susceptibility over time.
Agent
Susceptible
Amoxicillin
High — maintained
Co-amoxiclav
High — maintained
Cefuroxime, cefpodoxime, ceftriaxone
High — maintained
Macrolides
33%
Cefaclor
28.7%
Penicillin
23.4%
Haemophilus influenzaeLocal data only
Susceptibility 2015–17, CLSI breakpoints.
Agent
Susceptible
Co-amoxiclav
100%
Levofloxacin
77.1%
Moxifloxacin
75.4%
Co-trimoxazole
41%
Penicillin at 23.4% does not mean amoxicillin fails. Reading those two lines together is the commonest misinterpretation of a pneumococcal sensitivity report. Amoxicillin retained high susceptibility in the same dataset in which penicillin sat at 23.4%. The gap is a breakpoint artefact — CLSI applies stricter thresholds to oral penicillin than to amoxicillin at adequate dose in non-meningeal infection. Do not abandon amoxicillin because the penicillin line looks bad.
What this adds up to
Amoxicillin, co-amoxiclav and the later cephalosporins held up. Amoxicillin at adequate dose remains a reasonable oral backbone.
Macrolides did not, at 33%. That removes both macrolide monotherapy and the reflexive azithromycin add-on unless there is a specific reason for atypical cover.
Co-trimoxazole at 41% against H. influenzae is not an empiric option.
Respiratory fluoroquinolones held up better against H. influenzae (75–77%) than fluoroquinolones did against urinary E. coli — but they remain a drug class worth reserving.
Step 5 · What to actually start
The algorithm, plus what it leaves to your judgement
Step 6 · Cultures, stewardship, and the system around you
Pakistan Chest Society, March 2026 · AMR country data
What a Pakistani audit found doctors actually doing Local data only
39%
of CAP patients received empiric therapy that did not match the guideline — 2022 multi-centre audit.
2.5×
higher mortality in patients who received inappropriate empiric treatment.
42%
of cases ran longer than the recommended duration.
25%
reduction in inappropriate antibiotic use where antimicrobial stewardship programmes were implemented, alongside shorter hospital stays.
The guideline lists four markers of inappropriate empiric therapy: macrolide monotherapy persisting despite resistance above 30%, common overuse of fluoroquinolones, empiric regimens not adjusted on culture or clinical response, and treatment failure, prolonged stays and ICU transfers linked to initial inadequate therapy.
Cultures, and why to send them even when the yield is low Pakistan-adapted
Empiric regimens are often not adjusted based on culture or clinical response.
That line sits in the guideline's list of what goes wrong. The value of a culture in Pakistan is not only the individual result — it is that the national susceptibility picture this guideline rests on is thin precisely because cultures are sent inconsistently, and thinner still outside Karachi, Lahore and Islamabad. Sending them in admitted patients, and acting on them at 48–72 hours, is how the next edition gets better numbers.
Why it is flagged: The cost-benefit of routine culture shifts when national surveillance data is scarce and resistance is changing fast.
The guideline's conditional additions
MRSA suspected: add vancomycin 15 mg/kg IV twice daily or linezolid 600 mg IV twice daily.
Pseudomonas suspected: avoid ceftriaxone and use an anti-pseudomonal beta-lactam — piperacillin-tazobactam or cefepime.
Both are conditional on a reason to suspect the organism. Neither is an empiric default, and treating them as one is how carbapenems and vancomycin stop working.
Three things about antibiotics in Pakistan that no international guideline accounts for
Antibiotics are widely available without prescription
The Drug Act of 1967 prohibits over-the-counter sale, but enforcement is inadequate — throughout Pakistan it is possible to obtain antibiotics without a legitimate prescription, generally on a simple request. Most classes are sold this way. Assume prior antibiotic exposure that the patient may not report as such, and ask specifically.
Pakistan is the third-highest antibiotic-consuming low- and middle-income country
That consumption is the engine behind the susceptibility figures on this page. Every avoidable prescription makes the next empiric choice worse.
Substandard and falsified products are a real confounder
Antibiotics account for 16.9% of all products reported as substandard or falsified in Pakistan. A patient who has genuinely failed a correct antibiotic at a correct dose may have received an inadequate product rather than harboured a resistant organism.
Several internationally recommended antibiotics are simply unavailable here
That is stated explicitly as a barrier to following international guidelines. A pathway that recommends a drug you cannot obtain is not a pathway.
Common questions
Can I use azithromycin for pneumonia in Pakistan?
Not on its own. The Pakistan Chest Society's March 2026 guideline states: avoid macrolide monotherapy due to high resistance, macrolide resistance above 30% in S. pneumoniae. Azithromycin still has a place as the second agent alongside a beta-lactam in admitted patients — what it is no longer is a drug you send an outpatient home on by itself.
What is first-line for an outpatient with pneumonia in Pakistan?
Amoxicillin 1 g three times daily, or doxycycline 100 mg twice daily, for 5 to 7 days. Co-amoxiclav 625 mg three times daily where there are mild comorbidities, or cefpodoxime 200 mg twice daily for 5 days.
Didn't the Pakistan Chest Society guideline say amoxicillin or a macrolide?
The previous edition did. The March 2026 edition replaced the macrolide option with doxycycline and ruled out macrolide monotherapy by name. If you learned the older guideline, this is the change to know.
Who counts as being at risk of a drug-resistant pathogen?
Hospitalisation in the past 90 days, antibiotics in the past 3 months, chronic lung disease, structural lung disease, chronic kidney disease or malignancy, tube feeding or aspiration risk, or known colonisation with a multidrug-resistant organism. Two or more of these call for extended-spectrum or dual therapy — Pakistani surveillance puts drug-resistant organisms in 40 to 45% of hospitalised CAP patients.
Penicillin susceptibility is only 23.4%. Does that mean amoxicillin will not work?
No, and this is the commonest misreading of a pneumococcal report. Amoxicillin retained high susceptibility in the same dataset. The gap is a breakpoint artefact — CLSI applies stricter thresholds to oral penicillin than to amoxicillin at adequate dose in non-meningeal infection.
Why be cautious with respiratory fluoroquinolones?
The guideline itself says avoid them in patients at risk of QT prolongation or with a cardiovascular history, and names fluoroquinolone overuse as a marker of inappropriate empiric therapy. On top of that, Pakistan has one of the world's highest tuberculosis burdens and fluoroquinolones have antimycobacterial activity — an empiric fluoroquinolone can partially treat undiagnosed TB, blunt the picture, delay diagnosis and select for fluoroquinolone-resistant TB.
Can I use CRB-65 instead of CURB-65?
Yes. CRB-65 omits serum urea, and the guideline names it as particularly useful in rural Pakistani setups and Basic Health Units without immediate lab access. Whichever you use, oxygen saturation below 90%, confusion or a respiratory rate of 30 or more should trigger escalation regardless of score.
How long should treatment run?
Outpatients 5 to 7 days if stable and improving; inpatients 7 to 10 days depending on severity. Courses of 5 days or fewer are reserved for mild to moderate disease with procalcitonin guidance and rapid recovery — rarely an option in Pakistan, where procalcitonin is seldom available. The guideline's own audit found duration exceeded recommended limits in 42% of cases.
Clinically reviewed by Dr Mehdi Raza, MBBS
MBBS — FCPS Postgraduate Trainee at Combined Military Hospital (CMH) Multan. Graduated from Federal Medical & Dental College, Islamabad. Full profile →
Sources
Pakistan Chest Society — Clinical Practice Guidelines: Community Acquired Bacterial Pneumonia, March 2026. "Prepared by the Pakistan Chest Society for National Guidance on the Diagnosis and Management of Community Acquired Bacterial Pneumonia." The source of every severity criterion, risk factor, regimen and dose on this page. Read from the published PDF, which is a scanned document with no text layer.
This page previously ran on the earlier, undated Pakistan Chest Society CAP guideline, whose outpatient first line was "amoxicillin or a macrolide". It was rebuilt on 6 August 2026 when the March 2026 edition was found. The earlier edition is superseded and should not be used.
Country data on AMR in Pakistan in the context of community-acquired respiratory tract infections: links between antibiotic susceptibility, local and international antibiotic prescribing guidelines, access to medicine and clinical outcome. Source of the S. pneumoniae and H. influenzae susceptibility figures (2015–17, CLSI breakpoints), the IDSA 2019 macrolide threshold, and the access, over-the-counter and substandard-medicine data.
The 2026 guideline's own resistance statements, for comparison: macrolide resistance in S. pneumoniae "exceeds 25%, which is the threshold used in international guidelines to discourage monotherapy"; in H. influenzae, ampicillin resistance 18–25%, macrolide resistance around 15–20%, and co-amoxiclav and cefuroxime above 95% susceptibility.
Background reading — what changed when the macrolide came out of first line, the breakpoint artefact that makes people abandon amoxicillin, and why the fluoroquinolone question is really about TB: pneumonia in Pakistan.