VOL.11 NO.5 MAY 2006 Medical Bulletin

17

VOL.13 NO.12 DECEMBER 2008

IntroductionPneumonia is the 3rd leading cause of death in HongKong, and it accounts for more than 4000 deaths peryear locally.1 It can be categorised into community-acquired or hospital-acquired, and it can affectimmuno-competent or immuno-compromised hosts.In this review, we will focus on the management ofcommunity-acquired pneumonia (CAP) in animmuno-competent host.

DefinitionPneumonia is an acute lower respiratory tract infectionassociated with fever, symptoms and signs in thechest, and abnormalities on the chest x-ray.Community-acquired pneumonia is defined as thepresence of symptoms and signs which are consistentwith an acute lower respiratory tract illness associatewith new radiographic shadow for which there is noother explanation in a patient who has not recentlybeen hospitalised.2

AetiologiesCommon bacterial pathogens causing CAP includeStreptococcus pneumoniae, Haemophilus influenzae,Moraxella catarrhalis, Mycoplasma pneumoniae,Chlamydophila pneumoniae and Legionella pneumophila.In Hong Kong, tuberculosis may masquerade as CAPin about 12% of patients.3 Influenza and otherrespiratory viruses (e.g. respiratory syncytial virus,adenovirus) can be complicated by viral pneumonia orsecondary bacterial pneumonia. In patients withchronic medical illness and/or alcoholism, Klebsiellapneumoniae may cause a fulminant pneumonia.Fulminant bacteraemic Acinetobacter baumannii CAP isa newly described entity which may occasionally causerapidly fatal pneumonia in Hone Kong and othertropical countries.4

Patient with travel history or unusual exposure mightcontract unusual organisms. A thorough historytaking is most important to provide clues to theseunusual agents (Table 1).

Clinical and Radiological FeaturesCommon clinical features of CAP include cough, fever,sputumproduction, dyspnoea and pleuritic chest pain.

Other features include gastrointestinal symptoms andmental state changes. General examination revealsfebrile, tachyponea or tachycardia. Examination of thechest detects localised crackles or bronchial breath sound.Chest radiograph (CXR) is helpful in confirming thediagnosis of pneumonia.

Radiographic changes include lobar consolidation, eitherunilobar or multilobar; interstitial infiltrates, cavitation,with or without pleural effusion. However, the pattern ofradiological involvement is seldom predictive of thelikely aetiology in CAP.

Assessment of SeverityCAP presents to physicians both in primary andsecondary care settings as a wide spectrum of illness,from mild and self-limiting to life-threatening or evenfatal disease. The decision regarding out-patienttreatment or in-patient care is the first and single mostimportant decision for general practitioners.

Different criteria and models have been developed togive an objective assessment of the severity ofpneumonia and to predict the mortality of pneumonia.They can assist in identifying patients with CAP whomay be candidates for out-patient treatment. The twomost commonly used prediction rules are thePneumonia Severity Index (PSI) and CURB-65.

Management of CommunityAcquired PneumoniaDr. Judy LamMBBS, MRCP (UK), FHKCP, FHKAM (Med), Specialist in Respiratory MedicineSpecialist, Division of Respiratory Medicine, Department of Medicine & Geriatrics,United Christian Hospital, Hong Kong

Dr. CMChuMD, MSc (Respirat Med) (Lond), FRCP (Lond, Edin, Glasg), FHKCP, FHKAM (Med),Specialist in Respiratory MedicineConsultant Physician, Department of Medicine & Geriatrics, United Christian Hospital, Hong Kong

Dr. CM ChuDr. Judy Lam

Table 1. Epidemiological clues to community-acquired pneumonia

Epidemiological linksAir conditioning, coolingtowers, travel and hotel,mist machine, hospitalCattle, sheep, goatsWindstorm in endemic area(e.g. California)Homeless, prisonMilitary camp

Nursing home

Bat cavesTurkeys, chickens, ducks,birdsRabbitsHealth care worker

Thailand/SE AsiaAlcoholism

OrganismsLegionella pneumophila

Coxiella burnetiiCoccidioides immitis

TuberculosisStreptococcus pneumoniae,Chlamydophila pneumoniae,adenovirus,Mycoplasma penumoniaeInfluenza A or B, Chlamydophilapneumoniae, Respiratory SyncytialVirus, Streptococcus pneumoniaeHistoplasma capsulatumInfluenza A, Chlamydophila psittaci

Francisella tularensisTuberculosis, HIV, Influenza A or B,SARSBurkholderia pseudomalleiStreptococcus pneumoniae, Klebsiellapneumoniae, Staphylococcus aureus,anaerobes

Medical Bulletin

18

VOL.13 NO.12 DECEMBER 2008

PSI is a prognostic model which uses demographics, co-morbidities, findings on physical examination andlaboratory findings to stratify patients into 5 mortalityrisk classes. Class I and II patients should be treated asout-patient, risk class III patients should be treated in anobservation unit or with a short hospitalisation, and riskclass IV and V patients should be treated as in-patients.5

CURB-65 is a severity assessment score which stratifiespatients into different mortality groups suitable fordifferent management pathways. One point is given foreach of following factors: Confusion (based on a specificmental test or disorientation to person, place, or time),Urea >7mmol/L (20mg/dL), Respiratory rate　30 breaths/min, low Blood pressure (systolic < 90mmHg or diastolic　60mmHg), and age　65 years.6 The higher the score,the higher the mortality risk (Score 0, 0.7%; Score 1, 5.3%;Score 2, 13%; Score 3, 17%, Score 4, 41.5%; Score 5, 57%).This score is adopted by the updated BTS guideline as aseverity assessment model (Table 2). Patients withCURB-65 score of 0-1 can be treated as out-patients, thoseof score 2 should be considered for a short stay in-patienttreatment or hospital supervised out-patient treatment,and patients with the score　3 should be managed inhospital and may need ICU care.7

ComplicationsPleural effusion occurs in 40% of bacterial pneumonia.8Usually the parapneumonic effusion is small andresolves with appropriate antibiotics. However, it canbecome complicated if there is persistent bacterialinvasion of the pleural space (which is indicated by apH of <7.2 and/ or LDH >1000 IU/l). Empyema thoracisdevelops if the bacteria are not cleared, which ischaracterised by bacterial organisms seen on Gram stainor positive culture, or aspiration of pus fromthoracentesis.

Lung abscess formation is another complication ofpneumonia. It is commonly caused by aspirationpneumonia (e.g. in alcoholic or debilitated patients), orless commonly, due to bacteraemia or tricuspid valve

endocarditis with septic emboli to the lung. This can bediagnosed by chest radiograph or more sensitively, byCT scan.

TreatmentIn Hong Kong, Streptococcus pneumoniae is the singlemost common pathogen in CAP. Therefore, the choicesof empirical therapy need to take into account of thelocal data of drug-resistant S. pneumoniae (DRSP). Also,the role of atypical pathogens in CAP (e.g. Chlamydiapneumoniae, Mycoplasma pneumoniae and Legionellaspecies) is increasingly recognised, patients should becovered for the possibility of these infections. Lastly, theplace of starting the antibiotic (out-patient or in-patient)and the presence of modifying factors, including riskfactors for DRSP, enteric Gram negatives andPseudomonas aeruginosa are also important factors toconsider for the choice of antibiotic.

The following recommendations are based on the localguideline IMPACT.9 For out-patient treatment of CAP,empirical therapy should cover the usual organisms,which includes an oral amoxicillin-clavulanate orampicillin-sulbactam macrolide or amoxicillin + a newermacrolide. For in-patient treatment, the preferredregimen includes intravenous or oral amoxicillin-clavulanate or ampicillin-sulbactam　macrolide. Thealternative treatment includes cefotaxime or ceftriaxone　macrolide. In serious CAP which requires ICUadmission, the treatment options should also coverEnterobacteriacea; choices include intravenouspiperacillin-tazobactam or cefotaxime or ceftriaxone +macrolide; another alternative is cefepime + macrolide(Table 3). If a patient is at risk of Pseudomonas infection,an anti-Pseudomonal antibiotic should be prescribed.

Fluoroquinolone is not recommended as the first linetherapy for CAP in Hong Kong. The resistance of thisclass of drug is rapidly emerging among the S.pneumonia in this locality.10 Moreover, tuberculosis isprevalent in Hong Kong. Excessive use offluoroquinolone may lead to delay in diagnosis oftuberculosis and increased fluoroquinolone resistanceamongMycobacterium tuberculosis.11,12

Besides antibiotics, supportive treatment should begiven as appropriate. Oxygen therapy should be givenif the patient is hypoxic, those with volume depletionmay require intravenous fluid. In those with prolongedillness, nutritional support is important.

Table 2. Severity assessment of CAP using CURB-65 score,adopted from BTS guideline for management of communityacquired pneumonia in adults - 2004 update

Table 3. Recommendation for the empirical therapy ofcommunity-acquired pneumonia according to IMPACT (3rdedition)

Confusion*Urea > 7mmol/lRespiratory rate　30/minBlood pressure (SBP < 90mmHg or DBP　60mmHg)Age　65 years

Score 1 point for each feature present

*Defined as a Mental Test Score of 8 or less, or new disorientation in person,place or time.

The Abbreviated Mental Test (each question scores 1 mark, total 10 marks):Age, date of birth, time (to nearest hour), year, hospital name, recognition oftwo persons, recall address, date of First World War, name of monarch, countbackwards 20　1

CURB-65Score

Likely suitablefor hometreatment

0 or 1 3 or more2

Considerhospitalsupervisedtreatment

CAPCAP, nothospitalised

CAP, hospitalizedin general ward

CAP, hospitalizedin ICU for seriouspneumonia

Usual organismsS. pneumoniaeH. influenzaeC. pneumoniaeC. psittaci(influenza A,M. tuberculosis)

As above

As above +Enterobacteriaceae

Preferred regimensPO Amoxicillin-clavulanate orampicillin-sulbactam　a macrolideorPO amoxicillin + anewer macrolideIV/ PO Amoxicillin-clavulanate orampicillin-sulbactam　a macrolidesIV Piperacillin-tazobactam orcefotaximeor ceftriaxone + amacrolide

Alternatives

Cefotaxime orceftriaxone　amacrolides

Cefepime + amacrolide

Manage inhospital asseverepneumoniaAssess for ICUadmissionespecially ifCURB-65 score= 4 or 5

VOL.11 NO.5 MAY 2006 Medical Bulletin

19

VOL.13 NO.12 DECEMBER 2008

The presence of empyema thoracis requires early andeffective drainage with chest tubes. In a local study,Streptococcus milleri, Bacteroides and Klebsiella pneumoniaewere the most common cultured organisms.13 The choiceof antibiotic in these cases should cover Gram negativeand anaerobic organisms as well. If the above treatmentsfail, surgical drainage is required. Lung abscesses usuallyrespond to a prolonged course of antibiotics. Lobectomyis seldom required.

Approach to Unresolved PneumoniaWhen the pneumonia is not responding to treatment,careful analysis of the history, physical examination,radiology and laboratory features is required. There aremany possibilities of unresolved pneumonia. It can bedue to inadequate dosage or coverage of antibiotics,resistant or unusual organisms, underlyingimmunodeficiencies, persistent predisposition ofpneumonia (e.g. foreign body, obstructing tumour,aspiration), development of complications (empyemathoracis, lung abscess, metastatic infection), orconcomitant processes (e.g. line sepsis, drug fever,Lemierre's syndrome, transfusion reaction, pulmonaryembolism, pseudomembranous colitis) causing fever.Finally, some diseases can also cause similar symptoms,signs and radiographic changes, mimicking pneumoniae.g. drug reaction, eosinophilic pneumonia, extrinsicallergic alveolitis, bronchiolitis obliterans organisingpneumonia (BOOP), diffuse alveolar damage (DAD),vasculitides, lipoid pneumonia, connective tissue diseaseand neoplasm. CT thorax, bronchoscopy and/or openlung biopsy may be required to differentiate the causes.

PrognosisWith treatment, most CAPs resolve within days toweeks. However, in older patients, presence of co-existing disease, high CURB-65 score, hypoxaemia,multilobar involvement, bacteraemia, fulminantorganisms and development of complications areassociated with adverse outcomes.

ConclusionWith the correct choice of antibiotic, a generalpractitioner can safely manage patients with low riskCAP in the out-patient setting. A follow up CXR isnecessary to confirm resolution of pneumonia. However,in high risk patients (Table 2) and in non-respondingcases, early referral for specialist care is advised.

Dr. James CM HoCentre for Health Protection website http://www.chp.gov.hk/.British Thoracic Society Guidelines for the management ofcommunity acquired pneumonia in adults. Thorax 2001; 56(supplIV):iv1-iv64.Chan CH, Cohen M, Pang J. A prospective study of community-acquired pneumonia in Hong Kong. Chest 1992; 101:442-446.Leung WS, Chu CM, Tsang KY, Lo FH, Lo KF, Ho PL. Fulminantcommunity-acquired acinetobacter baumannii peumonia as a distinctclinical syndrome. Chest 2006; 129:102-109.Fine MJ, Auble TE, Yealy DM, et al. A prediction rule to identify lowrisk patients with community-acquired pneumonia. N Eng J Med1997; 336:243-250.Lim WS, MM van der Eerden, et al. Defining community acquiredpneumonia severity on presentation to hospital: an internationalderivation and validation study. Thorax 2003; 58: 377-382.British Thoracic Society Guidelines for the management ofcommunity acquires pneumonia in adults - 2004 updates.Light RW, Girard WM, Jenkinson SG, et al. Parapneumoniceffusions. Am J Med 1980; 69:507.Reducing bacterial resistance with IMPACT - Interhospital Multi-disciplinary Programme on Antimicrobial ChemoTherapy. Thirdedition.Ho PL, Yung RWH, Tsang DN, Que TL, Ho M, Seto WH et al.Increasing resistance of streptococcus pneumoniae to fluoroquinolones:results of a Hong Kong multi-centre study in 2000. J AntimicrobChemother 2001.Sterling TR. The WHO/IUATLD diagnostic algorithm fortuberculosis and empiric fluoroquinolone use: potential pitfalls. Int JTuberc Lung Dis 2004; 8: 1396-1400.Dooley KE, Golub J, Goes FS, Merz WG, Sterling TR. Empirictreatment of community-acquired pneumonia withfluoroquinolones, and delays in the treatment of tuberculosis. ClinInfect Dis 2002; 34:1607-1612.Tsang KY, Leung WS, Chan VL, Lin AWL, Chu CM. Complicatedparapneumonic effusion and empyema thoracis: microbiology andpredictors of adverse outcomes. Hong Kong Med J 2007; 13:178-86.

1.2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

References