View metadata, citation and similar papers at core.ac.

uk brought to you by CORE

provided by OAR@UM

Concordance Study

Hospital management of community-acquired

pneumonia in Malta

Roberta Callus, Josef Micallef, Jonathan Mamo, Stephen Montefort

Abstract Introduction
Community-acquired pneumonia (CAP) remains a common Community-acquired pneumonia (CAP) is a common and
diagnosis requiring hospital admission and a leading cause potentially life threatening infectious disease with significant
of death worldwide. No local guideline is currently available impact on patient morbidity and mortality as well as hospital
for the management of CAP. Our aim was to evaluate current resources. According to population based studies, the annual
practices in the management of CAP at Mater Dei Hospital, incidence rate of radiologically proven CAP in adults varies
Malta. In this prospective study we looked at all adult patients between 2.6 to 13.4 per 1000 inhabitants, with an increased
admitted with CAP in winter and summer (105 consecutive days incidence in males and at the extreme ages of life.1 Up to 50%
for both seasons). Data collected and analysed included: basic are reported to require hospital admission. In 2007, 1.2 million
patient demographics; symptoms at presentation; antibiotics people in the USA. were hospitalised with pneumonia.2 Critical
care admissions with CAP vary from 5% in the British Thoracic
prescribed and time of administration; co-morbidities; CURB65;
Society (BTS) multicentre study3 to 10% in a Spanish study.4
blood oxygenation levels; admission plan; length of stay and
Globally, pneumonia causes more deaths than any other
follow-up; length of stay, follow-up chest radiography and death
infectious disease. In 2007 more than 52,000 people died from
within 30 days from hospital admission. Of note the average time
the disease in the USA.2 Mortality rates in hospitalised patients
to first dose antibiotic was 7 hours 48 minutes (range 2 hours 13
ranges from 5-15% in various developed countries.1 Patients
minutes – 14 hours 17 minutes). A total of 178 patients (50.1%)
requiring intensive care have a mortality of up to 55%.5 Mortality
were admitted with CURB65 scores of 0 (n=99) and 1 (n=86).
increases with age and in the presence of co-morbidities.6 The
Most of these could have been discharged and managed in the CURB65 score (confusion; urea >7mmol/L; respiratory rate ≥ 30
community with significant impact on hospital bed occupancy. breaths/min; blood pressure <90 mmHg systolic or ≤ 60mmHg
Eventual standardisation of acute management of CAP by the diastolic; aged >65 years old) is a verified mortality prediction
set-up of a local guideline will improve outcome and reduce tool adopted by the BTS to assess the severity of CAP.
hospital bed occupancy. The management of CAP may vary both within and in
between countries for a number of reasons including different
lung pathogens, in or out-patient management, caring physician,
antibiotic availability and different health-care systems.7
For this reason several guidelines have been published by
various international societies to ensure standardisation in
management.
Keywords The aim of this study was to evaluate current practices in
Pneumonia, mortality, management audit the management of CAP in the Accident and Emergency and
Internal Medicine departments at Mater Dei Hospital, Malta.

Methods
CAP was defined as an acute respiratory illness acquired
Roberta Callus* MD, MRCP(UK) outside hospital with characteristic symptoms and signs
35, Vjal il- Bon Pastur Balzan BZN 1621 associated with the appearance of new radiographic infiltrates
Email: [email protected] on chest radiography. Patients from nursing homes and those
with a diagnosis of aspiration pneumonia, tuberculosis or
Josef Micallef MD, MRCP(UK)
[email protected] doubtful initial radiographic appearances were excluded. All
adult patients (above 16 years) with CAP admitted to Mater
Jonathan Mamo MD Dei Hospital, Malta were included in a prospective study. The
[email protected] two periods studied were: December 1, 2008 to March 15, 2009
Stephen Montefort FRCP, PhD (winter) and June 1, 2009 to September 13, 2009 (summer).
[email protected] The picture archiving and communication systems
(PACS) system was used to review radiological findings.
*corresponding author

6 Malta Medical Journal Volume 24 Issue 02 2012

Admission and clinical data were collected from the patients’
medical case notes. The data collected included: symptoms at Figure 2: Percentage of recovered vs deceased in
presentation; antibiotics prescribed and time of administration; relation to target time to antibiotic administration
co-morbidities; CURB65; blood oxygenation levels (oxygen
saturation and/or arterial partial pressure of oxygen); admission
plan; length of stay (LOS), follow-up chest radiography and
death within 30 days from hospital admission.
The CURB65 score was used to assess severity on
admission.8 A co-morbidity score was devised to include the
following risk factors: smoking, diabetes mellitus, chronic
kidney disease, cardiovascular disease, respiratory disease
and immunosuppression (malignancy, HIV, patients on long-
term steroids or immunosuppressants). Each co-morbidity
was a given a score of 1. The total score ranged from 0 to a
maximum of 6. Ethical approval was obtained from the Malta
University Research Ethics Committee after all consultants in
the department of medicine gave their written consent.
Antibiotic use
Thirty one percent of admitted patients (n=110) had received
Statistical analysis
at least one dose of antibiotic at home prior to hospitalisation. The
The data were analysed using Statistical Package for Social
main antibiotic combination prescribed in hospital was intravenous
Sciences (SPSS) version 16.0. We used chi-square to analyse
cefuroxime and oral clarithromycin (38.6%, n=137), followed by
associations between categorical variables and t-test for
intravenous co-amoxiclav and oral clarithromycin (13.5%, n=48).
continuous variables.
Monotherapy with co-amoxiclav, cefuroxime or clarithromycin
was prescribed in 6.8% (n=24). Fluoroquinolone monotherapy
Results
was prescribed in 10.3% (n=37). More potent antibiotics were
Incidence and demography
prescribed in 30.8% (n=109), of whom 26.1% (n=29) had no
A total of 355 patients with CAP were included during the
co-morbidities. There was a change in the original antibiotic
study periods: 193 and 162 patients during winter and summer
prescription in 57.2% throughout the admission (n=203).
respectively. This gives an annual inpatient incidence rate of
The average time to first dose antibiotic was 7 hours 48
1.62 cases per 1,000 inhabitants (95%CI 1.42 to 1.82). The mean
minutes (range 2 hours 13 minutes – 14 hours 17 minutes). There
daily number of general medical admissions was 46, of which
was 73.2% (n = 260) of the whole cohort who received antibiotics
1.7 were admitted with CAP. The mean age was 65 (range 16-
outside the four-hour target. From the deceased patients, 75%
95). Sex distribution was 182 males and 173 females. Twenty
(n = 39) had received their first antibiotic dose after four hours
patients (5.6%) required ICU care and 24 (6.8%) were admitted
(Figure 2). The average total antibiotic treatment duration was
to a high dependency medical ward.
7.2 days.

Symptoms at presentation
Co-morbidity score
The two main presenting symptoms were cough (75.8%)
The mean co-morbidity score was 1.8. This increased to 2.4
and fever (66.2%), followed by shortness of breath (54.4%) and
in the deceased patients (p = 0.049).
sputum expectoration (53.5%). Pleuritic chest pain was present
in only 16.1%. Most patients had more than one symptom at
CURB65
presentation (Figure 1).
The CURB65 score was used to assess severity on admission.
In the study population 52% (n=184) of admissions scored
CURB65 of 0 (28%, n=99) or 1 (24%, n=85). The most common
CURB65 score at presentation was 2 with 110 (31%) patients.
Figure 1: Symptoms at presentation Twelve percent (n=43) and 5% (n=18) had a CURB65 score of
3 and 4 respectively.
Increasing CURB65 scores showed positive correlation with
30-day mortality in both seasons. Patients admitted during
the winter period with a CURB65 score of 1, 2 and 4, had a 1.7,
1.4 and 1.6 increased risk of death within 30 days respectively,
despite similar co-morbidities (Figure 3).

Malta Medical Journal Volume 24 Issue 02 2012 7

 Outcome: 30-day mortality
Figure 3: Seasonal 30-day mortality in relation to The average 30-day mortality for the whole population
CURB65 + under study was 14.6% (n=52). There was a difference between
the winter and summer cohorts: 17.6% (n=34) versus 11.1%
(n=18) respectively (p=0.085). The mean age of the deceased
was 78 years (range 51-95). Patients admitted to ICU had 15%
mortality (n=3) and all died in winter. High dependency medical
ward admissions had 25% mortality: 36.4% (n=4) winter, 15.4%
(n=2) summer.

Respiratory/Infectious Disease firms versus

General Medical firms
General medical firms admitted 56.3% (n=200) of patients
and respiratory and infectious disease firms admitted 43.7%
(n=155). An independent-sample t-test was conducted to
compare length of hospital stays (LOS) between General Medical
firms and Respiratory firms. There was no significant difference
for General Medical (M = 7.96, SD = 6.79) and Respiratory and
Infectious Disease firms (M = 7.40, SD = 5.89; t (354) = 0.83,
p=0.408, two-tailed). The magnitude of the differences in the
Oxygen prescription means (mean difference = 0.56, 95% CI: -0.77 to 1.89) was very
Oxygen saturations were assessed with pulse oximetry and/ small (eta-squared = 0.0006). General medical firms had a 30-
or arterial blood gas sampling. Oxygen saturation on air was day mortality of 15.5% whilst respiratory and infectious disease
measured in 85.1% (n=302) patients and on oxygen in 39.4% firms had 13.6%.
(n=140). Oxygen was prescribed with a PaO2 of ≤60mmHg in
52.4% (n=186). Discussion
CAP remains a common presentation in the A&E department
Further management of every hospital. Decisions whether hospital or outpatient
Blood cultures and sputum cultures were requested in management is most appropriate are important in determining
47.1% (n=167) and 60.8% (n=216) respectively. C-reactive outcome, hospital stay and bed occupancy.9 In May 2005
protein (CRP) was taken in 15.8% (n=56). Chest physiotherapy the European Respiratory Society in collaboration with the
was requested in 54.1% (n=192). Thrombo-prophylaxis was European Society of Clinical Microbiology and Infectious
prescribed in 29.2% (n=104) (Figure 4). Diseases (ESCMID) published new guidelines for the
management of adult lower respiratory tract infections.10
Follow up chest radiography The incidence of CAP is very variable in most European
A chest radiograph was repeated during admission in 16.3% countries ranging from 11 per 1000 in Finland11 to 1.6 per 1000
(n=58) and after discharge in 19.2% (n=68). This was repeated in Spain and Italy.12 For the period under study, the incidence
in 8.7% (n=31) both during admission and upon discharge. No rate was 1.62 per 1000 which is similar to neighbouring
follow-up chest radiograph was done in 67.3% (n=204). Mediterranean countries. This excludes patients managed in
the community by general practitioners.
In the USA there were 17.3 deaths per 100,000 with an in-
Figure 4: Further management patent mortality of 5.4% in 2006.2 The local 30-day mortality
of 14.6% was relatively high when compared to other European
countries, however in a recent study the 30-day mortality in a
UK GP cohort was 18.5% in the community.13 The most widely
studied scoring systems are the CURB65 and the Pneumonia
Severity Index (PSI). Each has its advantages and limitations
with the CURB65 easily identifying more severely ill patients and
the PSI identifying low mortality risk patients. The limitation
of these prognostic tools includes their variable utility in
stratifying risk in the elderly, associated co-morbidities and
social factors. The use of such scoring systems, combined
with clinical judgement is crucial for determining the need for
hospitalisation.10 This was clearly demonstrated in this study.
Mortality remains high in CURB65 score of 2 or more. These
patients will require hospitalisation, intensive treatment and

8 Malta Medical Journal Volume 24 Issue 02 2012

monitoring. A total of 178 patients (50.1%) were admitted with Sputum cultures are another controversial laboratory tool.
CURB65 scores of 0 (n=99) and 1 (n=86). Most could have been In our study these were requested in 60.8%, however the actual
discharged and managed in the community with significant number reaching the laboratory was much less and most were
impact on hospital bed occupancy. This study shows a 1.7- after first antibiotic administration. Samples should be sent to
fold increased risk of 30-day mortality in winter. This may be the laboratory in moderate to severe CAP and in those failing
due to: more aggressive pathogens, later presentation, longer to improve clinically. The Infectious Disease Society of America
waiting times, time to effective treatment and inpatient care in (IDSA) and the Canadian Infectious Disease Society/Canadian
a busy hospital. It is difficult to say which is the most important Thoracic Society (CIDS/CTS) recommend routine sputum
contributing factor, but all could play a role. analysis for all inpatients with CAP, while the American Thoracic
The antibiotic choice should be guided by illness severity, Society (ATS) recommends this only if a drug-resistant pathogen
patient age, co-morbidities, clinical presentation, epidemiology, or an organism not covered by usual empiric therapy is suspected.
previous antibiotic exposure and local antibiotic resistance.14 CRP whose initial description was based on pneumonia
Antibiotic choice and time to first dose of administration remain patients is useful for establishing the severity of CAP.22 Seppa et al.
issues of discussion.15 International guidelines currently aim for reported that a CRP level >100 mg/L is a marker independently
administration within four to six hours of admission, which was associated with higher risk of death.23 A Scandinavian study
not achieved in our hospital. This was due to long waiting times observed that patients with higher CRP levels had longer fever
and lack of antibiotic access within our A&E department, which duration, longer hospital stays and fewer patients had recovered
were then administered on the wards. Blood cultures and waiting clinically or radiographically at 8 weeks follow-up.24 Failure of
for the next scheduled drug round also delayed administration. CRP to fall by 50% or more at day 4, independent of admission
The most commonly prescribed combination was cefuroxime CRP level, was shown to be associated with increased 30-day
and clarithromycin. Respiratory fluoroquinolones were mortality and need for mechanical ventilation and/or inotropic
prescribed mostly in patients with penicillin allergy. Locally support and complicated pneumonia.25 In our study CRP was
almost 30% of Streptococcus pneumoniae strains are resistant only taken in 15.8% mostly on the day after admission. Another
to macrolides, 16 however these antibiotics cover atypical biomarker gaining interest is procalcitonin (PCT). PCT rises in
organisms. Appropriate and judicious antibiotic use should response to systemic inflammation associated with infection and
be guided by local sensitivity and resistance patterns keeping correlates well with pneumonia severity.26,27 A low PCT (<0.228
in mind tolerability and toxicity in each individual patient. ng/ml) has a high negative predictive value for mortality from
Treatment directed towards particular pathogens suspected CAP when compared with the CRB65 score.26
clinically is highly commendable. Chest physiotherapy has been widely used as a routine
The duration of antibiotic treatment is usually 7 to 10 adjunctive therapy for pneumonia. A review of six randomised
days, except for intracellular pathogens (Legionella sp.) where controlled trials concluded that chest physiotherapy has no
more prolonged treatment is necessary.17 The current trend benefit in reducing LOS, and mortality during the acute phase
is to further shorten antibiotic treatment.18 Switching from of CAP.28
intravenous to oral treatment depends on clinical resolution Guidelines for prevention of venous thromboembolism
and ability to tolerate oral intake. This allows earlier hospital recognise pneumonia as a risk factor causing significant
discharge reducing costs and risk of hospital-acquired infections. morbidity and mortality. Decreased mobility, dehydration and
There is considerable variability in LOS between hospitals co-morbidities increase this risk further. This study reflects local
reflecting variations in clinical practice preferences, hospital inadequacy in prescribing thromboprophylaxis.
characteristics and patient characteristics and attitudes.19 The According to the BTS, the chest radiograph need not be
mean LOS varies in many countries from 7 days in the USA20 to repeated prior to hospital discharge in those with a satisfactory
11.7 days in Spain.1 Our local mean LOS was 7.7 days. clinical recovery. A chest radiograph should be arranged after
The BTS states that all patients should receive appropriate about 6 weeks for patients with persistence of symptoms or
oxygen therapy with the aim of maintaining PaO2 >8 kPa and physical signs or those at higher risk of malignancy (smokers
SaO2 >92%. Oxygenation levels should be assessed by pulse and age >50 years). Our study shows that most of the discharged
oximetry, followed by arterial blood gas analysis if oxygen patients (n=204) were not followed up with a chest radiograph.
saturation is <92% and in patients with chronic obstructive
lung disease. Only half of the patients in this study were Conclusion
prescribed appropriate oxygen therapy. Oxygen therapy should Standardisation of acute management of CAP will improve
be clearly prescribed in the treatment chart including flow rate, outcome and reduce hospital bed occupancy. A clear delineation
concentration and mask type. of hospital admission criteria on basis of severity scores is
A number of studies have questioned the clinical value and essential as a significant number of patients may be managed in
cost-effectiveness of routine blood cultures.21 Many clinicians the community with the establishment of a general practitioner/
agree that these are important in severely ill patients (CURB65 nurse-led home treatment programme. Timely and appropriate
≥2 or PSI IV, V) and if a specific risk factor for pathogens antibiotics, supportive treatment and laboratory tools may
resistant to empirical therapy is present. optimise inpatient management. Admission under respiratory

Malta Medical Journal Volume 24 Issue 02 2012 9

or infectious disease firms may also improve outcome. Clinical 13. Myles PR, Hubbard RB, Gibson JE, Pogson Z, Smith CJ, McKeever
TM. Pneumonia mortality in a UK general practice population
improvement together with radiology and laboratory tools
cohort. Eur J Public Health. 2009 Oct;19(5):521-6.
reflect resolution of CAP and hence decision on discharge. 14. Mandell LA, Bartlett JG, Dowell SF, File JM, Musher DM, Whitney
Follow up with a chest radiograph in six weeks is often necessary. C. Infectious Disease Society of America update of practice
guidelines for the management of community-acquired pneumonia
The set up of a local guideline on the management of CAP is in immunocompetent adults. Clin Infect Dis. 2003;37:1403-33.
pivotal for providing excellence in clinical care of these patients. 15. Pines JM, Isserman JA, Hinfey PB. The measurement of
time to first antibiotic dose for pneumonia in the emergency
department: A white paper and position statement prepared for the
References American Academy of Emergency Medicine. J Emerg Med. 2009
1. Almirall J, Bolibar I, Vidal J, Sauca G, Coll P, Niklasson B et al. Oct;37(3):335-40.
Epidemiology of community-acquired pneumonia in adults: a 16. National Antibiotic Committee Malta. Antibiotic resistance – facts
population-based study. Eur Respir J, 2000;15:757-63. and figures. Nov 2010.
2. Centers for Disease Control and Prevention - Pneumonia. 17. Mandell LA, Bergeron MG, Gribble MJ, Jewesson PJ, Low DE,
[Accessed Dec 22 2010]. Available from: http://www.cdc.gov/ Marrie TJ, et al. Sequential antibiotic therapy: effective cost
features/pneumonia. management and patient care. Can J Infect Dis. 1995;6:306.
3. British Thoracic Society Research Committee and Public Health 18. el Moussaoui R, de Borgie CA, van den Broek P, Hustinx WN,
Laboratory Service. The aetiology, management and outcome of Bresser P, van den Berk GE, et al. Effectiveness of discontinuing
severe community-acquired pneumonia on the intensive care unit. antibiotic treatment after 3 days versus 8 days in mild to moderate-
Respir Med. 1992;86:7-13. severe CAP: randomised, double blind study. BMJ. 2006;332:1355-
4. Torres A, Serra-Batlles J, Ferrer A, Jiménez P, Celis R, Cobo E, 61.
et al. Severe community-acquired pneumonia. Epidemiology and 19. Menendez R, Cremdes MJ, Martinez-Moragon E, Soler JJ,
prognostic factors. Am Rev Respir Dis. 1991;144:312-8. Reyes S, Perpina M. Duration of length of stay in pneumonia:
5. Lee JH, Ryu YJ, Chun EM, Chang JH. Outcomes and prognostic influence of clinical factors and hospital type. Eur Respir J. 2003
factors for severe community-acquired pneumonia that requires Oct;22(4):643-8.
mechanical ventilation. Korean J Intern Med. 2007;22:157-63. 20. Fine MJ, Medsger AR, Stone RA, Marrie TJ, Coley CM, Singer DE,
6. Waterer GW, Kessler LA, Wunderink RG. Medium-term survival et al. The hospital discharge decision for patients with community-
after hospitalization with community-acquired pneumonia. Am J acquired pneumonia. Arch Intern Med. 1997;157:47-56.
Respir Crit Care Med. 2004;169(8):910-4. 21. Campbell SG, Marrie TJ, Anstey R, Dickinson G, Ackroyd-Stolarz
7. Woodhead M. Community-acquired pneumonia guidelines S. The contribution of blood cultures to the clinical management of
– an international comparison: a view from Europe. Chest. adult patients admitted to the hospital with community-acquired
1998;113(3):183S-7S. pneumonia. Chest. 2003;123:1142–50.
8. Lim W, van der Eerden MM, Laing R, Boersma WG, Karalus N, 22. Luna CM. C-reactive protein in pneumonia: let me try again. Chest
Town GI, Lewis SA, and Macfarlane JT. Defining community- 2004;125(4):1192–5.
acquired pneumonia severity on presentation to hospital. Thorax. 23. Seppä Y, Bloigu A, Honkanen PO, Miettinen L, Syrjälä H..
2003;58(5):377-82. Severity assessment of lower respiratory tract infection in elderly
9. Fine MJ, Hough LJ, Medsger AR, Li WH, Ricci EM, Singer DE. The patients in primary care. Arch Intern Med 2001;161:2709-13.
hospital admission decision for patients with community-acquired 24. Hedlund J. Diagnostic and prognostic value of interleukin-6 and
pneumonia. Arch Intern Med. 1997;157:36-44. C-reactive protein in community-acquired pneumonia. Scand J
10. Woodhead M, Blasi F, Ewig S, Huchon G, Ieven M, Ortqvist Infect Dis 1995;27(5):457–62.
A, et al; European Respiratory Society; European Society of 25. Chalmers JD, Singanayagam A, Hill AT. C-reactive protein is
Clinical Microbiology and Infectious Diseases.Guidelines for the an independent predictor of severity in community-acquired
management of adult lower respiratory infections. Eur Respir J. pneumonia. Am J Med. 2008;121:219–25.
2005;26:1138-80. 26. Christ-Crain M, Stolz D, Bingisser R, Muller C, Miedinger D,
11. Jokinen C, Heiskanen L, Juvonen H, Kallinen S, Karkola K, Huber PR, et al. Procalcitonin guidance of antibiotic therapy
Korppi M, et al. Incidence of community-acquired pneumonia in in community-acquired pneumonia: a randomised trial. Am J
the population of four municipalities in Eastern Finland. Am J Respir Crit Care 2006;174:84-93.
Epidemiol. 1993;137:14-8. 27. Masiá M, Gutiérrez F, Shum C, Padilla S, Navarro JC, Flores E,
12. Almirall J, Morató I, Riera F, Verdaguer A, Priu R, Coll P, et al. et al. Usefulness of procalcitonin levels in community-acquired
Incidence of community-acquired pneumonia and Chlamydia pneumonia according to the patients outcome research team
pneumoniae infection. Eur Respir J. 1993;6:14-8. pneumonia severity index. Chest. 2005;128:2223.
Yang M, Yuping Y, Yin X, Wang BY, Wu T, Liu GJ, et al. Chest
physiotherapy for pneumonia in adults. Cochrane Database Syst
Rev 2010;2:CD006338.

10 Malta Medical Journal Volume 24 Issue 02 2012