Pleural infection, empyema and lung abscess

• Pleural infection is bacterial invasion and replication in the pleural space, usually arising from an adjacent pneumonia but sometimes without any pneumonia on imaging. Empyema is the end stage, with frank pus in that space. ^[1][2]
• A parapneumonic effusion moves through three stages - exudative, fibrinopurulent, and organising - and the stage dictates the treatment: antibiotics alone, then antibiotics plus drainage, then possible surgery. ^[2][6]
• Drainage is needed once the fluid shows an empyema, a low pH of 7.2 or under, or other signs of complicated infection. ^[1][2]
• The standard intrapleural rescue regimen is tissue plasminogen activator (tPA) 10 mg plus deoxyribonuclease (DNase) 5 mg, each twice daily for 3 days, for fluid that remains after the first drain. Single-agent therapy should not be used. ^{[1][2] [7]}
• A lung abscess is a cavity in the lung parenchyma, most often from aspiration of mouth organisms. Most resolve with prolonged antibiotics; cavities larger than 6 cm and abscesses that fail medical therapy may need drainage or surgery. ^[4]

By the end of this page you should be able to:

1. Define pleural infection and empyema, and distinguish a simple from a complicated parapneumonic effusion.
2. Explain the three stages of parapneumonic effusion from exudate to fibrothorax, and link each to management.
3. Use pleural fluid pH, glucose, lactate dehydrogenase (LDH), Gram stain and imaging to decide when to insert a chest drain.
4. Outline empirical antibiotic regimens for community-acquired and hospital-acquired pleural infection, and the realistic duration of treatment. ^[6]
5. Describe intrapleural tPA and DNase dosing, safety, and how it compares with surgery.
6. Recognise the RAPID score, what it predicts, and its limits.
7. Diagnose an aspiration lung abscess and manage it with antibiotics, drainage and surgery, including doses.
8. Recognise complications and prognosis of both conditions, and plan care in a low-resource setting.

Much of the terminology in this area is used loosely, and the labels matter because they imply different actions. A parapneumonic effusion is any effusion accompanying pneumonia or a lung abscess. An uncomplicated parapneumonic effusion is free-flowing and settles with antibiotics alone. A complicated parapneumonic effusion (or complicated pleural infection) will not resolve without drainage: classical fluid markers are pH under 7.2, glucose under 40 mg/dL (2.2 mmol/L), LDH over 1000 IU/L, or a positive Gram stain or culture. Empyema is pus, a positive Gram stain or culture, and is an absolute indication for drainage. ^{[2][6] [1]}

The European Respiratory Society statement defines pleural infection more biochemically, as a collection with infective symptoms plus a low pH under 7.2, a low glucose under 2.2 mmol/L, or culture positivity, and states that empyema is frank pus. The British Thoracic Society (BTS) defines it mechanistically as bacterial entry and replication in the cavity, and takes a multimodal diagnostic approach. Guidelines therefore genuinely disagree on where the threshold for infection is set, and the two staged definitions are both in routine use. ^[2][3]

A parapneumonic effusion and an empyema belong on a spectrum with pleural infection, whereas a lung abscess is a cavity in the lung itself. The link between them is aspiration and anatomy: a lung abscess can seed the pleura and create an empyema, and a deep adjacent infection can produce both at once. If a parenchymal cavity and an effusion coexist, the fluid must be sampled, because empyema behaves differently from abscess and needs drainage. ^[4][6]

Pleural infection is becoming more common worldwide. Hospitalisation for empyema in United States adults rose from 8.1 to 11.1 per 100,000 between 2007 and 2016. Cases predominantly affect men, with an average age near 60 years, although adults under 65 years make up 42 to 60% of cases. Common comorbidities are cancer, diabetes, cardiovascular disease, renal failure and alcohol use; in one large review, a median of 72% of patients had at least one comorbidity. An ageing population, more immunosuppression, and wider use of computed tomography and bedside ultrasound - which find collections earlier - all plausibly contribute. ^[2][6]

Outcomes are serious. A 6-year Australian study of culture-positive cases reported inpatient mortality of 13.4% for community-associated and 16.6% for healthcare-associated pleural infection, rising to 32.4% and 45.5% respectively after one year. The PILOT study reported 10% mortality at 3 months and 19% at 12 months. Hospital stay averages 2 to 3 weeks. Median survival differences are driven more by the patient's premorbid state than by the microbe in most series. ^[2][6]

Lung abscess epidemiology is less precisely quantified. Historically, empyema complicated about 5% of pneumonias before antibiotics and about 2% afterwards; current series report primary abscesses in otherwise healthy people and secondary abscesses in those with dental disease, alcohol use, aspiration risk, airway obstruction or immunosuppression. ^[6][4]

Two routes are recognised. Direct spread from an adjacent pulmonary focus is the classic one, and it creates parapneumonic infection. Haematogenous or aspiration-seeding routes probably account for many primary pleural infections where no pneumonia is seen, since molecular studies often recover oral organisms such as Fusobacterium nucleatum and Streptococcus intermedius, and the pleural microbiome looks more like the mouth than the lung. Iatrogenic introduction is the third route and is rare: infection after an indwelling pleural catheter occurs in 2.5 to 6% of cases. ^[2][6]

The infection is polymicrobial in a substantial minority. In a meta-analysis of over 10,000 isolates, Staphylococcus aureus was the commonest single organism at 20.7%, followed by viridans streptococci (18.7%), Pseudomonas (17.6%), Enterobacteriaceae (11.9%), Streptococcus pneumoniae (10.8%), Klebsiella (10.7%) and coagulase-negative staphylococci (4.5%); 12.1% were anaerobes and 12.9% were polymicrobial, 75% of those including an anaerobe. Community-acquired cases are more often streptococcal; hospital-acquired cases more often yield Staphylococcus aureus, including methicillin-resistant strains in roughly two thirds, plus Gram-negative organisms such as E. coli, Enterobacter and Pseudomonas. Where S. pneumoniae or S. anginosus is isolated, the prognosis tends to be better; S. aureus and Gram-negatives carry poorer outcomes. ^[2][6]

Conventional pleural fluid cultures are positive only in 31 to 58% of cases, more often when fluid is inoculated into blood culture bottles, which increases yield by about 20%. Molecular methods such as 16S ribosomal RNA sequencing reach positivity of 82 to 93% and often reveal anaerobes that culture misses. Because organisms are missed so often, microbiology alters treatment less often than one would hope, but source control and early broad coverage remain essential. ^[2][6]

Risk factors for pleural infection are the same as for severe pneumonia: older age, diabetes, chronic kidney disease, cancer, immunosuppression, chronic lung disease, poor oral hygiene and alcohol use. A higher comorbidity burden raises mortality. Roughly three quarters of hospitalised patients with pleural infection undergo tube thoracostomy, video-assisted thoracoscopic surgery (VATS) or open thoracotomy. ^[2][6]

Most primary abscesses follow aspiration of oropharyngeal contents, especially during reduced consciousness, dysphagia, gastro-oesophageal reflux, vomiting or heavy alcohol use, so anaerobes dominate: Bacteroides, Prevotella, Peptostreptococcus, Fusobacterium and streptococci. Poor dental hygiene is an independent risk factor, because it increases the anaerobic load. Secondary abscesses follow bronchial obstruction by tumour, foreign body or lymph node, pre-existing lung disease such as bronchiectasis or bullous emphysema, cystic fibrosis, infected infarct, or haematogenous spread from infective endocarditis, abdominal sepsis or septic thromboembolism. Klebsiella pneumoniae, Staphylococcus aureus, Pseudomonas, Nocardia and Actinomyces appear more in secondary and immunocompromised cases. ^[4]

Aspiration-related cavities settle in gravity-dependent segments: the posterior segment of the right upper lobe, the right middle lobe, or the superior segment of a lower lobe. Poor dentition, alcohol use, seizures, bulbar dysfunction and cognitive impairment are the main population risk factors. Secondary abscesses carry a poor prognosis - roughly 75% mortality - largely because they mark underlying malignancy or immunosuppression. ^[4]

The usefulness of the stage concept is that each stage has a matching treatment, and late stages are harder and more morbid to fix. ^[2][6]

1. Exudative stage. Infection in the adjacent lung inflames the visceral pleura, raising capillary permeability. Protein- and neutrophil-rich fluid accumulates, but it is usually sterile on culture and Gram stain at this point: glucose over 60 mg/dL and LDH less than three times the upper limit of serum normal. This is an uncomplicated parapneumonic effusion and usually resolves with antibiotics alone. ^[6]
2. Fibrinopurulent stage. Bacteria invade the space, recruiting neutrophils and activating coagulation. Fibrin is deposited, septa form, and fluid becomes loculated. Markers worsen: pH falls under 7.2, glucose falls, LDH rises, and the fluid is turbid or purulent. Drainage is now needed. ^[2][6]
3. Organising stage. Fibroblasts proliferate into a dense fibrous peel covering the visceral and parietal pleura, driven by growth factors such as transforming growth factor beta and platelet-derived growth factor. The lung becomes trapped, producing fibrothorax and non-expandable lung. Antibiotics alone are no longer enough; decortication may be needed. ^[2][6]

The transition is a time-dependent failure of source control: inadequate or delayed antibiotics let inflammation progress through fibrin deposition into fibrosis. Imaging already hints at it - loculations, septations, enhancing split pleura and thickened extrapleural fat are the fingerprints of the later stages. ^[2][6]

Primary abscesses begin as aspiration pneumonia. Unrecognised, the pneumonitis progresses to tissue necrosis within 1 to 2 weeks, a cavity forms, and a bronchopulmonary communication creates an air-fluid level. Polymicrobial sepsis and polymicrobial acidic products produce the characteristic foul, putrid sputum. Secondary abscesses follow obstruction with post-obstructive pneumonia, or haematogenous seeding forming multiple smaller, usually basal, cavities. ^[4]

Presentation is often that of pneumonia - fever, cough, dyspnoea and pleuritic chest pain - but an insidious course with weight loss, night sweats and anaemia is also common, and a lack of radiographic pneumonia is reported in about 56% of community-acquired and 73% of hospital-acquired pleural infection. Physical examination shows the usual signs of effusion: reduced expansion, stony dull percussion and reduced breath sounds. Small collections may be silent except on imaging, so any non-resolving pneumonia, or fever with inflammatory markers plus an effusion, should raise suspicion. ^[2][6]

Early symptoms mimic pneumonia - fever, chills, cough, night sweats, chest pain and dyspnoea - with a productive cough developing once the cavity communicates with a bronchus; sputum may be purulent, foul-smelling and blood-streaked. Anaemia and weight loss are common in chronic disease. A cavitating lesion with thick irregular walls, absence of fever or purulent sputum, and lack of leucocytosis should prompt consideration of squamous cell or small cell lung cancer rather than infection. ^[4]

Imaging comes first. Chest radiography detects an effusion (blunting of the costophrenic angle at roughly 200 mL on a posteroanterior film, 50 mL on a decubitus film) and may show the cause. ^[7] Lung ultrasound is the bedside tool of choice: it detects effusions under 50 mL, distinguishes septated from free-flowing collections, and guides thoracentesis and drain insertion. ^[7] Computed tomography of the thorax defines pleural and parenchymal disease, loculations, the split pleura sign, enhancement and fat stranding. A collection over 30 mL and split pleura together help distinguish complicated from simple parapneumonic effusion before any tap. ^[2][6] Ultrasound and CT detect septations better than chest radiography, and these are themselves associated with the need for drainage. Magnetic resonance imaging of the pleura remains research-level but is feasible and avoids ionising radiation.

Thoracentesis should be done as soon as pleural infection is suspected if there is a safe volume. Send fluid for appearance, paired protein and LDH, cell count and differential, pH, glucose, Gram stain and culture (aerobic and anaerobic, ideally inoculated into blood culture bottles, which raise yield by about 20%), and consider cytology and tuberculosis tests where relevant. Measure pH immediately with a blood gas analyser, before local anaesthetic or heparin contaminates the sample. ^[1][2][6]

Pleural fluid pH is the primary marker: under 7.2 means a high probability of pleural infection and mandates a drain. In the intermediate range of 7.2 to 7.4, drain if LDH exceeds 900 IU/L, especially with other supportive features - ongoing fever, large volume, glucose at or below 4.0 mmol/L (72 mg/dL), CT enhancement or septations on ultrasound. With no pH available, an initial glucose under 3.3 mmol/L (about 60 mg/dL) is a useful proxy. Loculations, turbidity, low glucose and pleural CRP above 10 mg/dL all support infection; pleural procalcitonin does not. ^[1][2]

Contrast-enhanced computed tomography is the investigation of choice and should be obtained when abscess is suspected on a plain film. It confirms the cavity and its wall, defines its relation to the pleura, and helps separate abscess from empyema, bullae, cysts and malignancy. Sputum examination and culture (plus tuberculosis and fungal cultures when indicated), blood cultures and, in immunocompromised patients or those with suspected endocarditis, echocardiography round out the work-up. Bronchoscopy is used to obtain microbiological material and to exclude obstruction by tumour or foreign body. Abscess distribution follows gravity: right upper lobe posterior segment, right middle lobe, or superior segment of a lower lobe. ^[4]

The three parapneumonic stages described above are the practical staging, because each maps to a management step: uncomplicated exudative disease gets antibiotics alone; fibrinopurulent disease needs drainage plus antibiotics; organising disease may need decortication. ^[2][6]

The Royal College of Physicians RAPID score (renal function, age, fluid purulence, infection source, and serum albumin) stratifies adults at presentation into low (0 to 2), intermediate (3 to 4) and high (5 to 7) risk groups; the high-risk group carries a 3-month mortality of about 30%. It predicts mortality and length of stay, but has not been shown to guide treatment and should not be used alone to select patients for surgery or intrapleural therapy. ^[1][2]

Assessment is clinical rather than scored. Small abscesses under about 2 cm are really necrotising pneumonia; an abscess is a cavity over 2 cm. Many abscesses smaller than about 6 cm resolve with antibiotics alone, but the evidence in sources only establishes that cavities larger than 6 cm are unlikely to resolve with antibiotics alone and may need drainage. Poor prognostic factors include older age, severe comorbidities, immunosuppression, bronchial obstruction and neoplasm. Empyema sometimes masquerades as parenchymal abscess, so any failure to respond, or a pleural fluid collection, must prompt sampling. ^[4]

Resuscitate first. Give oxygen for hypoxaemia, fluids or vasopressors for shock, and treat sepsis. A massive or tension-like collection may require urgent drainage. In uncomplicated parapneumonic effusion, antibiotics alone usually suffice. For complicated or empyema disease, source control through a chest drain is the cornerstone of therapy. ^[2][6]

Choice of drain matters less than technique and timing. Small-bore tubes (14 French or smaller) are at least as effective as large-bore drains and less painful, based on the MIST-1 data and a subsequent meta-analysis. ^[1][2] Place the drain under ultrasound or CT guidance. Repeated therapeutic thoracentesis is reasonable in selected stable patients; for an indwelling tube, keep the drainage rate to 500 mL per 8-hour shift or 1.5 L per day to reduce re-expansion pulmonary oedema. ^[7] Drain removal thresholds genuinely differ between current sources: under 200 mL a day for two consecutive days in one 2026 review ^[7], and under 50 mL in 24 hours or imaging showing lung re-expansion in StatPearls Thoracic Empyema. ^[8]

All guidelines recommend prompt empirical broad-spectrum antibiotics as soon as pleural infection is suspected, chosen by infection setting, local resistance, and pleural penetration, with cover for anaerobes in almost all cases because the culture yield is poor. ^[3][6]

• Community-acquired infection. Target Gram-positive aerobes and anaerobes. Reasonable regimens are a second or third-generation cephalosporin, or an aminopenicillin with a beta-lactamase inhibitor (for example ampicillin-sulbactam), each combined with an anaerobe-directed agent such as metronidazole; clindamycin is an alternative where appropriate. Penicillin alone covers S. pneumoniae and H. influenzae, but mixed infection is common enough that anaerobic cover is justified. ^[2][6]
• Hospital-acquired infection. Cover resistant Gram-negatives, including Pseudomonas aeruginosa, plus MRSA, with broader antipseudomonal therapy such as cefepime-metronidazole, piperacillin-tazobactam or a carbapenem, plus vancomycin or linezolid, and metronidazole for anaerobes. Staphylococcus aureus comprises about 50% of positive cultures in this setting, two thirds of them methicillin resistant. ^[2][6]
• Penicillin allergy. In penicillin allergy, current pleural-infection reviews describe substitution with a quinolone (moxifloxacin or levofloxacin) plus anaerobic cover such as metronidazole; practice tables also use clindamycin with or without a quinolone in some summaries. ^[2]
• Avoid aminoglycosides. They penetrate the pleura poorly and are inactivated in the acidic infected fluid. ^[6]
• Agents that penetrate well. Penicillin, ceftriaxone, metronidazole and clindamycin achieve pleural levels equivalent to blood; linezolid exceeds blood levels in infected pleural fluid. ^{[6] [2]}

Duration is the least evidence-based part of management. Consensus is 2 to 6 weeks, guided by clinical response, inflammatory markers and imaging, with an initial intravenous course of 5 to 7 days commonly described in recent reviews (for example Kenig 2026) to settle fever and markers before oral step-down once the patient is afebrile, stable and absorbing. Shorter courses (14 to 21 days versus 28 to 42 days) showed similar efficacy in the SLIM trial, and 2 weeks was non-inferior to 3 weeks in a small trial of amoxicillin-clavulanate. Early switch from intravenous to oral treatment is appropriate when drainage and source control have been achieved; prolonging therapy after surgical decortication does not clearly add benefit. Post-operative courses after VATS were comparable short (median 11.6 days) and long (median 29.1 days) in one study. ^{[2][3][6] [4]}

When drainage and antibiotics leave a residual collection or ongoing sepsis, combination intrapleural tPA and DNase is the established rescue. Monotherapy with either drug alone should not be used, and streptokinase should not be used. ^[1][2] The BTS position is that surgery is first-line unless the patient is unfit or surgical delay exceeds 48 hours; the European Respiratory Society task force would initiate tPA/DNase within 48 hours of standard care as a potential surgery-sparing option, with surgical referral run in parallel. These two guidelines genuinely disagree, and the reason given by reviewers is that no high-quality head-to-head comparison exists. ^[3]

The standard MIST-2 protocol is 10 mg alteplase instilled, followed by a 10 mL 0.9% saline flush, then 5 mg DNase, a further 10 mL saline flush, clamping for 1 hour, then free drainage, given twice daily for 3 days (6 doses). Lower starting doses of 5 mg or 2.5 mg tPA appear to work in most patients, with escalation if response is insufficient, and roughly 75% of respondents in an international survey of pleural experts reported having used lower doses. Concurrent administration of the two drugs also appears safe and effective. ^[1][2][3]

Bleeding is the serious complication. Clinically significant intrapleural bleeding occurred in 4.1% of 1,833 patients in a multicentre cohort, defined as pleural fluid haematocrit reaching half the blood haematocrit, or 25 to 50% with clinical suspicion prompting intervention. Predictors are therapeutic anticoagulation, a high RAPID score, elevated serum urea and platelets under 100 x 10^9/L. Anticoagulation should be withheld before intrapleural therapy where possible. The risk does not fall clearly with lower tPA doses, so the benefit of de-escalation is cost and nursing workload rather than proven safety. Saline irrigation (instilling 250 mL of 0.9% saline three times a day for 3 days) is an alternative when intrapleural therapy or surgery is unsuitable. Systemic corticosteroids do not help: the STOPPE trial found no benefit from dexamethasone. ^[2][3][6]

Timing of rescue is itself a judgement. European guidance suggests considering rescue within 2 to 3 days of drain insertion if there is still fever, failure of CRP to fall by more than half, or persistent effusion, rather than waiting a week. ^[2][6]

Surgical drainage and decortication are the definitive option for patients with ongoing sepsis or persistent effusion after drainage and intrapleural therapy, or for high-risk or unstable patients and late presenters. VATS should be considered over open thoracotomy, because hospital stay is shorter, though complication and mortality rates are similar and conversion to thoracotomy occurs in up to 8.4% of cases. Post-operative mortality for pleural infection ranges from 1% to 12%, varying by timing and type of surgery, with advanced age, chronic kidney disease, cancer, late surgery and post-operative pulmonary complications as risk factors. Extent of surgery should be tailored to empyema stage and tailored around drainage versus debridement when the lung is not completely trapped; decortication is an individual decision for a trapped lung. ^[1][3]

Early surgical intervention is associated with shorter length of stay and possibly lower failure rates. The evidence is mainly retrospective and subject to selection, so no validated threshold for how early defines benefit has been established. A 2017 American Association for Thoracic Surgery consensus advocated surgery as first-line for stage II empyema, so older surgical-first literature still reflects that practice. Two feasibility trials (MIST-3 and a US pilot) showed similar length of stay between surgery and intrapleural therapy, with a quality-of-life signal favouring intrapleural therapy, but definitive large trials are awaited. ^[2][3]

Repeat imaging by CT in follow-up is recommended after a pleural infection, mainly to exclude an occult malignancy. Empyema recurrence after adequate drainage is uncommon and late presentation of obstruction should prompt search for cancer or non-infectious disease. After discharge, antibiotics are continued to complete the planned course, with review of the organism if culture-positive. ^[1][2]

• Fibrothorax and trapped lung from the organising stage, which may need decortication months later. ^[2][6]
• Conversion of VATS to open thoracotomy in up to 8.4% of cases, and reoperation in 2.3 to 3.9%. ^[3]
• Intrapleural bleeding at 4.1% with combination therapy. ^[2][3]
• Surgery-related pulmonary complications, with the risk factors above, and post-operative mortality of 1% to 12%. ^[3]
• Progression to chronic infection, bronchopleural fistula and pleurocutaneous fistula. ^[4]
• Re-admission and healthcare costs are high, and pleural infection worsens the prognosis of the underlying pneumonia. ^[2]

Prognosis tracks the patient more than the organism. RAPID high-risk patients have roughly 30% 3-month mortality; culture-positive disease has higher 3-month mortality (about 25%) than culture-negative (about 10%); staphylococcal and Gram-negative infections and polymicrobial infections fare worse; S. pneumoniae, S. anginosus and anaerobes fare better. Prompt source control and early consideration of rescue therapy are the modifiers within clinician control. ^[2][6]

Prevention is mostly prevention of aspiration pneumonia and early drainage of parapneumonic effusion. Vaccination against S. pneumoniae and influenza reduces pneumonia burden. Good dental care, cautious alcohol use, swallowing assessment and positioning in at-risk patients reduce aspiration. Early ultrasound-guided drainage before fluid organises is itself the best prevention of chronic empyema. ^[4][6]

Most abscesses are diagnosed after a pneumonia that does not clear. Give empirical antibiotics promptly after taking sputum, blood and, where possible, bronchoscopic samples; do not delay treatment for a perfectly sterile diagnosis. Evaluate for obstruction: a new abscess in a smoker or an older patient, or one in a non-dependent segment, warrants bronchoscopy to exclude cancer or a foreign body. ^[4]

Beta-lactamase inhibitors are the preferred empiric therapy: ticarcillin-clavulanate, ampicillin-sulbactam, amoxicillin-clavulanate or piperacillin-tazobactam, followed by imipenem or meropenem if needed. Anaerobic cover is mandatory even when cultures are negative, because anaerobes are frequently present and hard to culture. ^[4]

A typical intravenous regimen is ampicillin-sulbactam 1.5 g (1 g ampicillin + 0.5 g sulbactam) to 3 g (2 g ampicillin + 1 g sulbactam) every 6 hours, with the total sulbactam dose not exceeding 4 g a day, adjusted for renal function (every 12 hours if creatinine clearance is 15 to 29 mL/min; daily in more severe impairment), given by slow intravenous injection over 10 to 15 minutes. ^[5]

Other agents:

• Penicillin-allergic patients. Clindamycin remains an alternative; its routine empiric use has declined because of Clostridioides difficile risk, and it is best reserved for those with true allergy when beta-lactams cannot be given. ^[4]
• MRSA coverage. Add vancomycin or linezolid if the patient has healthcare exposure within 3 months. Daptomycin has no activity against pulmonary infections. ^[4]
• Methicillin-sensitive Staphylococcus aureus. Cefazolin 2 g intravenously every 8 hours, nafcillin 2 g intravenously every 4 hours, or oxacillin 2 g intravenously every 4 hours, with renal adjustment. ^[4]
• Oral step-down. Once the patient is afebrile, stable and tolerating oral intake, amoxicillin-clavulanate is the oral agent of choice. ^[4]
• Metronidazole monotherapy is not useful because of polymicrobial flora. ^[4]

Duration is about 3 weeks, varying with clinical and radiological response, and cavities may take weeks to close; antibiotics should not be stopped at clinical improvement alone. Secondary and healthcare-associated abscesses need longer courses guided by clinical, radiological and microbiological response, always alongside treatment of the underlying condition. ^[4]

Antibiotics alone cure over 80 to 90% of primary abscesses. Reserve escalation for specific situations: ^[4]

• Abscess larger than 6 cm, or unlikely to resolve medically. ^[4]
• Clinical failure after several days of appropriate antibiotics, or persistent sepsis, haemoptysis, bronchopleural fistula, or empyema. ^[4]

Options, in increasing invasiveness, are bronchoscopic drainage or postural drainage, CT-guided percutaneous catheter drainage (small drainage tubes are now the mainstay after failed medical therapy, with high success rates), and surgery - lobectomy or pneumonectomy - for failed medical therapy in fit patients. For poor surgical candidates, percutaneous and endoscopic drainage is preferred. Drainage of a lung abscess is not a first-line intervention and should not be rushed: premature, unnecessary drainage increases complication risk. ^[4]

• Immunocompetent patients usually resolve in about 3 weeks. Secondary abscesses, particularly in immunocompromised patients or with bronchial neoplasm, carry a mortality of about 75%. ^[4]
• Pregnancy. Management is case by case and multidisciplinary; rare case reports have used antibiotics with or without lobectomy. ^[4]
• Complications. Rupture into the pleural space (empyema or pyopneumothorax), bronchopleural fistula, pleurocutaneous fistula, pleural fibrosis, trapped lung, respiratory failure. ^[4]

• Children. Lung abscess is less common; think of immune deficiency, cystic fibrosis, foreign body and congenital malformation. Pleural infection in children is a distinct entity and its guidelines differ. ^[4]
• Older adults. Comorbidity drives prognosis; secondary abscesses and hospital-acquired pleural infection carry the worst outcomes. ^[2][4]
• Immunocompromised. Higher rates of complications, opportunistic organisms (including fungi and Mycobacterium abscessus), and MRSA coverage are needed. ^[4]
• Pregnancy. Chest infection with abscess or empyema requires early imaging and multidisciplinary management; drugs and procedures should be chosen with obstetric input. ^[4]
• HIV and tuberculosis. Tuberculous effusion and abscesses present similarly but need anti-tuberculosis therapy; see the tuberculosis topic.

• Bedside ultrasound is the single highest-yield tool: it confirms effusion, finds septation, and makes thoracentesis and small-bore drain insertion safe. ^[2][6]
• Drain early. Small-bore tubes (14 French or smaller) with a Seldinger technique are acceptable and less painful; keep total drainage at or below 500 mL per 8-hour shift, or 1.5 L per day, to reduce re-expansion pulmonary oedema. ^[1][6][7]
• Antibiotics pragmatically. Use local guidelines and antibiograms. An aminopenicillin with a beta-lactamase inhibitor or co-amoxiclav, with metronidazole added, covers community infection; reserve broader therapy for suspected hospital infection. In low-resource settings, avoid adding aminoglycosides. ^[6]
• Without intrapleural fibrinolysis or surgery. Delayed surgical referral means prolonged courses; early referral to the nearest centre with VATS capability matters; antibiotics and drainage cure a large proportion of simple or uncomplicated empyema if started early. ^[1][6]
• Abscess care. Prolonged oral amoxicillin-clavulanate after IV induction, with postural drainage and good oral hygiene, cures most primary abscesses. CT is ideal but clinical and X-ray follow-up is often sufficient. ^[4]
• Tuberculosis mimic. In regions where TB is common, a lymphocyte-predominant effusion or cavitary upper-zone lesion should trigger TB testing before assuming pyogenic disease. ^[4][6]

• Pleural infection is bacterial replication in the pleural space; empyema is pus. It may occur with no visible pneumonia. ^[1][2]
• Parapneumonic effusion progresses exudative -> fibrinopurulent -> organising, and treatment escalates from antibiotics alone to drainage to possible decortication. ^[2][6]
• Drain if there is frank pus, a positive Gram stain or culture, pH 7.2 or under, or an intermediate pH with LDH over 900 IU/L plus supportive clinical or imaging features. ^[1][2]
• A glucose under 3.3 mmol/L (about 60 mg/dL) is a useful proxy when pH is unavailable; loculations on CT or ultrasound are themselves linked to the need for drainage. ^[1][2]
• Use a small-bore tube (14 French or smaller) placed under image guidance. ^[1][2]
• Intravenous antibiotics are given for the first 5 to 7 days in many published courses, then oral to complete 2 to 6 weeks, guided by clinical, biochemical and radiological response; anaerobic cover is recommended in almost all cases. ^[2][6]
• Aminoglycosides should not be used; they penetrate the pleura poorly and are inactive in acidic infected fluid. ^[6]
• Rescue for a residual collection is combined intrapleural tPA 10 mg plus DNase 5 mg, twice daily for 3 days; monotherapy should not be used. ^[1][2]
• Intrapleural bleeding occurs in about 4.1% of treated patients; anticoagulation should be withheld where possible. ^[2][3]
• BTS favours surgery first-line; the European Respiratory Society task force favours early tPA/DNase with parallel surgical referral. This disagreement is honest and reflects absent head-to-head trials. ^[3]
• VATS is preferred over open thoracotomy, with conversion in up to 8.4% and post-operative mortality of 1% to 12%. ^{[3] [1]}
• The RAPID score (0 to 2 low, 3 to 4 intermediate, 5 to 7 high, the last with 3-month mortality about 30%) predicts outcome but should not drive treatment choice. ^[1][2]
• An aspiration lung abscess is usually polymicrobial anaerobic; beta-lactamase inhibitors are first-line, with about 3 weeks of therapy and amoxicillin-clavulanate for oral step-down. ^[4]
• Abscesses over 6 cm, or failing to respond to appropriate therapy, may need bronchoscopic, percutaneous or surgical drainage. ^[4]
• Secondary abscesses and hospital-acquired pleural infection carry high mortality and need treatment of the underlying cause. ^[4][6]

Pleural effusion; community-acquired pneumonia; aspiration pneumonia; tuberculosis; malignant pleural effusion; lung cancer; bronchiectasis; infective endocarditis; swallowing disorders; respiratory failure and hypoxaemia; decortication and thoracotomy; antimicrobial stewardship.