• 
• 

Three Part Question

In [ child with pneumonia due to penicillin-resistant Streptococcus pneumoniae] does [treatment with penicillin alone] result in [higher morbidity or mortality]?

Clinical Scenario

You are looking after a previously healthy 3-year-old girl, who was admitted with radiographically confirmed lobar pneumonia 3 days ago. She was started on intravenous penicillin and remains clinically stable but continues to be febrile. The laboratory informs you that Streptococcus pneumoniae has grown from her blood culture and that the isolate is penicillin-resistant (minimum inhibitory concentration (MIC) 1 μg/ml). You are uncertain whether this has any implications for her clinical course and wonder whether you should change her antibiotic treatment on the basis of this information.

Search Strategy

Medline was searched via the OVID interface
Medline:(pneumococc* OR streptococcus pneumoniae) AND (pneumonia OR sepsis OR bacteraemia OR bacteremia OR septicaemia OR septicemia OR invasive disease) AND (penicillin-resistan* OR penicillin resistan* OR ampicillin-resistan* OR ampicillin resistan*). Limits set: human, English-language. This produced 917 matches, of which seven were relevant

Searches of additional databases using the same strategy, including the ISI Web of Science (1955 to present; 1108 matches), EMBASE (1980 to present; 671 matches) and Scopus (1950 to present; 358 matches), identified no further publications that had not already been found.

Search Outcome

All relevant publications were hand-searched for further references; no further relevant publications were identified. In instances where the original manuscript provided insufficient information, the authors were contacted and asked to provide further details. The search date was 24 May 2009.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Friedland et al, 1995	Children (age: 0–14 years) hospitalised with pneumococcal pneumonia (n = 78). Penicillin-susceptible: n = 53 (68%). Penicillin-resistant: n = 25 (32%). Definition of resistance intermediate: MIC >0.06–1.0 μg/ml, high-level resistance: MIC ≥2 μg/ml. Treatment: variety of antibiotic regimens; directed by attending physician	Prospective cohort study (level 1b)	Duration of fever, oxygen requirement and resolution of symptoms and signs	No statistically significant differences between the two groups in all outcome measures	Part of a larger study population of children with invasive pneumococcal disease. Nine cases in PRSP group received a cephalosporin and/or vancomycin; in three cases in the PRSP group treatment was changed following isolation of the causative organism 24% of the cases were HIV-infected (proportion significantly greater in PRSP group)
			Death	Number of deaths:PRSP: 4 (16.0%); PSSP: 4 (7.6%)*p = 0.26; OR 2.33 (95% CI 0.53 to 10.23)
Tan et al, 1998	Children (age: 0–25 years) with pneumococcal pneumonia (n = 254; number of episodes = 257). Penicillin-susceptible: n = 221 (86%). Penicillin-resistant: n = 36 (14%). Definition of resistance, intermediate: MIC 0.1–1.0 μg/ml (n = 22), resistant: MIC ≥2 μg/ml (n = 14). Treatment: variety of antibiotic regimens; directed by attending physician	Retrospective cohort study (level 2b)	Duration of fever, duration of oxygen requirement, need for chest drain, duration of hospitalisation	No statistically significant differences between the two groups in all outcome measures	More than 80% of the patients were treated with a 2nd or 3rd generation cephalosporin and/or vancomycin (exact figure unclear). “No difference” between antimicrobial regimen in PRSP and PSSP group according to the authors. The patient in the PRSP group with fatal outcome was treated with ceftriaxone and vancomycin.
			Death	Number of deaths:PRSP: 1 (2.8%); PSSP: 5 (2.3%)(note: all patients had underlying illness) *p = 0.60; OR 1.23 (95% CI 0.14 to 10.89)
			Duration of fever, duration of oxygen requirement, need for chest drain, duration of hospitalisation	No statistically significant differences between the two groups in all outcome measures
			Death	Number of deaths:PRSP: 1 (2.8%); PSSP: 5 (2.3%)(note: all patients had underlying illness) *p = 0.60; OR 1.23 (95% CI 0.14 to 10.89)
			Duration of fever, duration of oxygen requirement, need for chest drain, duration of hospitalisation	No statistically significant differences between the two groups in all outcome measures
			Death	Number of deaths:PRSP: 1 (2.8%); PSSP: 5 (2.3%)(note: all patients had underlying illness) *p = 0.60; OR 1.23 (95% CI 0.14 to 10.89)
Deeks et al, 1999	Children (age: 0–5 years) hospitalised with pneumococcal pneumonia (n = 87). Penicillin-susceptible: n = 52 (60%). Penicillin-resistant: n = 35 (39%). Definition of resistance intermediate: MIC 0.12–1.0 μg/ml (n = 12) high-level resistance: MIC ≥2 μg/ml (n = 23). Treatment: penicillin (100 000–200 000 units/kg/day) or ampicillin (dose not stated) monotherapy.	Retrospective cohort study (level 2b)	Favourable response (defined as blood cultures reverting to sterile or clinical improvement)	No statistically significant differences between the two groups RR for treatment failure 1.2 (95% CI 0.9 to 1.6) for intermediate and high-level resistance combined; RR 1.3 (95% CI 0.9 to 1.8) for high-level resistance alone	Study included further patients with pneumococcal pneumonia treated with other antibiotic regimens
			Need for supplemental oxygen, ICU admission, duration of hospitalisation Complications Death	No statistically significant differences between the two groups in all outcome measuresNo significant increase in complications in PRSP group (OR 1.0; 95% CI 0.6 to 1.7)†No significant increase in mortality risk in PRSP group (OR 1.1; 95% CI 0.3 to 3.8)†
Pirez et al, 2001	Children (age: 0–14 years) hospitalised with pneumococcal pneumonia (n = 51). Penicillin-susceptible: n = 30 (59%). Penicillin-resistant: n = 11 (22%). No susceptibility results: n = 10 (20%). Definition of resistance intermediate: MIC 0.12–1.0 μg/ml (n = 6), fully resistant: MIC ≥2 μg/ml (n = 5). Treatment: penicillin (200 000 units/kg/day) or ampicillin (300 mg/kg/day) monotherapy	Prospective cohort study (level 1b)	Complications (empyema, pneumatocele)	No statistically significant difference between both groups	Part of a larger study population of children with pneumonia. In 10 patients the diagnosis of pneumococcal infection was based on antigen detection (therefore no susceptibility data available)
			Death	Number of deaths:PRSP: 1 (9.1%); PSSP: 1 (3.3%)*p = 0.47; OR 2.9 (95% CI 0.17 to 50.85)
Hsieh et al, 2004	Children (age: 0–12 years) hospitalised with pneumococcal pneumonia (n = 50). Penicillin-susceptible: n = 2 (4%). Penicillin-resistant: n = 48 (96%). Definition of resistance intermediate: MIC 0.1–1.0 μg/ml (n = 29), resistant: MIC ≥2 μg/ml (n = 19). Treatment: variety of antibiotic regimens	Retrospective cohort study (level 2b)	Complications (empyema, necrotising pneumonia)	No statistically significant difference between groups infected with susceptible, intermediate or fully resistant pneumococci	Part of a larger study population of children with pneumoccocal pneumonia. Susceptibility testing only carried out in 50 cases All but 4 patients treated with 2nd or 3rd generation cephalosporin or amoxicillin combined with clavulanic acid
			Death	Number of deaths:PRSP: 2 (4.2%) (both intermediate strains); PSSP: 0 *p = 1.00; OR 0.27 (95% CI 0.01 to 7.24)
Wexler et al, 2006	Children (age: 0–16 years) with pneumococcal pneumonia (n = 111). Penicillin-susceptible: n = 84 (76%). Penicillin-resistant: n = 16 (14%). No susceptibility results: n = 11 (10%). Definition of resistance, intermediate: MIC 0.1–1.0 μg/ml (n = 8), fully resistant: MIC ≥1 μg/ml (n = 8). Treatment: “nearly all” patients treated with 2nd generation cephalosporin according to the authors	Retrospective cohort study (level 2b)	Duration of fever, pulmonary complications, duration of hospitalisation	No statistically significant differences between the two groups in all outcome measures	Susceptibility results available for 90% of the study population Pulmonary complications” included: effusion, empyema, atelectasis, pneumatocele and pneumothorax
Cardoso et al , 2008	Children (age: 0–5 years) hospitalised with severe pneumococcal pneumonia (n = 240). Penicillin-susceptible: n = 120 (50%). Penicillin-resistant: n = 120 (50%). Definition of resistance intermediate: MIC 0.12–1.0 μg/ml (n = 64), fully resistant: MIC ≥2 μg/ml (n = 56). Treatment: penicillin (200 000 units/kg/day) or ampicillin (150 mg/kg/day) monotherapy	Prospective cohort study (level 1b)	Treatment failure (defined as deterioration or persistence of fever, tachypnoea, dyspnoea or hypoxaemia)	No significant difference in the proportion of treatment failure between groups infected with susceptible, intermediate or fully resistant pneumococci (p = 0.75) Adjusted RR for treatment failure: intermediate: 1.14 (95% CI 0.58 to 1.97) resistant: 1.03 (95% CI 0.49 to 1.90)	Part of a larger study population of children with pneumoccocal pneumonia Patients in whom antibiotics were changed in the first 48 h were excluded Power calculation suggests that the study would have detected a difference in RR ≥2.5 (significance level of 0.05; power of 0.80)

Comment(s)

On a global scale, pneumonia is a leading cause of death in childhood. S pneumoniae is the most commonly isolated causative bacterial organism in community-acquired pneumonia, accounting for 9% to 55% of cases overall(Heffelfinger).

Penicillin-resistance in pneumococci was first described in the late 1960s. Since the 1990s the frequency of resistant strains has been rising rapidly, with rates of resistance varying significantly between different geographical regions(Heffelfinger). The most recent surveillance data from the UK indicate that currently approximately 6% to 8% of clinical, invasive pneumococcal isolates are penicillin resistant (Farrell).

As with other bacteria, resistance in pneumococci is categorised according to the inhibition of bacterial growth in vitro. The most widely used classification for penicillin-resistant S pneumoniae (PRSP) is that defined by the US National Committee for Clinical Laboratory Standards (now the Clinical and Laboratory Standards Institute, CLSI)as follows: (i) susceptible: MIC ≤0.06 μg/ml; (ii) intermediate resistance: MIC 0.12–1.0 μg/ml; and (iii) resistant: MIC ≥2.0 μg/ml. In 2008 the CLSI revised this definition, with classification of susceptibility now additionally being dependent on the site of isolation (ie, meningitis vs non-meningitis) and the mode of treatment (ie, oral vs intravenous)(CDC).

In many countries, penicillin or other β-lactam antibiotics, such as ampicillin or amoxicillin, remain the first-line treatment for pneumonia. The latest World Health Organization guidelines recommend the use of penicillin or ampicillin for the treatment of severe pneumonia in children under the age of 5 years (WHO). The most recent British Thoracic Society guidelines for the management of community-acquired pneumonia in childhood recommend amoxicillin as empiric therapy for non-severe pneumonia in children younger than 5 years of age and specifically state that “if clinical or microbiological data suggest that S. pneumoniae is the causative organism, amoxicillin, ampicillin, or penicillin alone may be used”(BTS).

There has been growing concern that the rising rates of PRSP may result in increasing treatment failure, which has led to ongoing debate about the appropriateness of such recommendations. In addition, there has been concern that penicillin-resistant strains may cause more severe disease than fully susceptible pneumococci(Heffelfinger).

Of the seven studies identified in our literature search, three were prospective cohort studies (Friedland, Pirez, Cardoso). while the remaining four were retrospective (Tan, Deeks, Hsieh, Wexler).In some reports, children with pneumococcal pneumonia represented only a subgroup within a larger study cohort of children with pneumonia (Pirez) or children with invasive pneumococcal disease (Friedland, Deeks). Although the definition for intermediate penicillin resistance varied between studies, all used an MIC of 1.0 μg/ml as the upper cut-off; all but one study (Wexler) defined high-level resistance as an MIC ≥2.0 μg/ml.

In four of the studies, patients were treated with variable antibiotic regimens, generally as directed by the attending physician; in all four studies a substantial proportion of patients received treatment with cephalosporins (Friedland, Deeks, Hsieh, Wexler). Consequently, these reports do not provide information regarding the effectiveness of penicillin. However, in all four reports there was no statistically significant difference between patients infected with PRSP and those infected with susceptible pneumococci regarding clinical course (including defervescence, duration of oxygen requirement, pulmonary complications and duration of hospital stay) and mortality. Another report (containing insufficient details for inclusion in the table) also suggested that complicated (compared with uncomplicated) pneumonia was no more common in children infected with PRSP (Tan).As early complications and mortality in pneumonia are closely associated with disease severity at presentation, these observations lend some support to the assumption that PRSP are not more virulent (Menendez).

In two of the three remaining studies, all patients received either penicillin or ampicillin as initial antibiotic treatment (Pirez, Cardoso).The study by Deeks et al, which included a subgroup of 87 patients with pneumococcal pneumonia receiving monotherapy with either penicillin or ampicillin, found that children infected with PRSP (intermediate and high-level combined) were statistically no more likely to fail treatment than children infected with susceptible organisms (RR 1.2; 95% CI 0.9 to 1.6). In addition there was no difference regarding the need for supplemental oxygen, admission to intensive care and the duration of hospitalisation. Pirez et al also reported that the rate of complications (empyema or pneumatocele) was not significantly increased in children infected with PRSP. The most recently published report was the prospective study by Cardoso et al, which also showed that children infected with either intermediate or highly resistant PRSP were statistically not more likely to experience treatment failure. In contrast to other studies, the authors provided a power calculation, which indicates that their study was large enough to detect a 2.5-fold difference between the groups.

The aforementioned studies provide reasonably strong evidence that infection with PRSP has no significant adverse impact on clinical course or mortality, and that treatment with penicillin or ampicillin monotherapy remains effective. This is consistent with data from several adult studies(Aspa, Pallares).

Most experts recommend using high doses of β-lactam antibiotics for the treatment of PRSP infections to ensure that adequate serum levels are achieved. The Drug-resistant Streptococcus pneumoniae Therapeutic Working Group recommends using intravenous benzylpenicillin at a dose of up to 300 000 units/kg (equivalent to approximately 180 mg/kg) per day (divided into four to six doses) or intravenous ampicillin at a dose of 100–300 mg/kg per day (divided into four to six doses)(Heffelfinger). These recommendations are supported by the data from the studies by Pirez et al and Cardoso et al, as well as pharmacokinetic data showing that with these doses serum concentrations exceeding 4 μg/ml are achieved for a sufficient duration between dosing intervals (Giachetto).

Two further aspects are important to emphasise. Firstly, it is generally agreed that penicillin monotherapy is inadequate for the treatment of children with PRSP infection of the central nervous system, including meningitis. Secondly, it currently remains unclear whether treatment with penicillin or ampicillin is adequate for invasive PRSP infections (including pneumonia) with strains with an MIC ≥4.0 μg/ml, as cases of treatment failure have been described in this setting (Pallares). In such cases, alternative therapy, such as a cefotaxime or ceftriaxone, is currently thought to be advisable (Heffelfinger). Notably, S pneumoniae strains with high-level resistance of this extent still remain uncommon in the UK (below 1%)(Farrell).

Editor Comment

CI, confidence interval; HIV, human immunodeficiency virus; ICU, intensive care unit; MIC, minimum inhibitory concentration; OR, odds ratio; PRSP, penicillin-resistant Streptococcus pneumoniae; PSSP, penicillin-susceptible S pneumoniae; RR, relative risk. *Calculated from data provided in the manuscript or by the author(s) using two-sided Fisher’s exact test; †based on the entire cohort of patients with invasive pneumococcal disease (rather than pneumococcal pneumonia in isolation); ‡includes additional data provided by the author(s) which were not reported in the original manuscript.

References

1. Friedland IR . Comparison of the response to antimicrobial therapy of penicillin-resistant and penicillin-susceptible pneumococcal disease. Pediatr Infect Dis J 1995;14:885–90.
2. Tan TQ, Mason EOJr, Barson WJ, et al . Clinical characteristics and outcome of children with pneumonia attributable to penicillin-susceptible and penicillin-nonsusceptible Streptococcus pneumoniae. Pediatrics 1998;102:1369–75.
3. Deeks SL, Palacio R, Ruvinsky R, et al . Risk factors and course of illness among children with invasive penicillin-resistant Streptococcus pneumoniae. The Streptococcus pneumoniae Working Group. Pediatrics 1999;103:409–13.
4. Pirez MC, Martinez O, Ferrari AM, et al . Standard case management of pneumonia in hospitalized children in Uruguay, 1997 to 1998. Pediatr Infect Dis J 2001;20:283–9.
5. Hsieh YC, Hsueh PR, Lu CY, et al . Clinical manifestations and molecular epidemiology of necrotizing pneumonia and empyema caused by Streptococcus pneumoniae in children in Taiwan. Clin Infect Dis 2004;38:830–5.
6. Wexler ID, Knoll S, Picard E, et al . Clinical characteristics and outcome of complicated pneumococcal pneumonia in a pediatric population. Pediatr Pulmonol 2006;41:726–34.
7. Cardoso MR, Nascimento-Carvalho CM, Ferrero F, et al . Penicillin-resistant pneumococcus and risk of treatment failure in pneumonia. Arch Dis Child 2008;93:221–5.
8. Heffelfinger JD, Dowell SF, Jorgensen JH, et al . Management of community-acquired pneumonia in the era of pneumococcal resistance: a report from the Drug-Resistant Streptococcus pneumoniae Therapeutic Working Group. Arch Intern Med 2000;160:1399–408.
9. Farrell DJ, Felmingham D, Shackcloth J, et al. Non-susceptibility trends and serotype distributions among Streptococcus pneumoniae from community-acquired respiratory tract infections and from bacteraemias in the UK and Ireland, 1999 to 2007. J Antimicrob Chemother 2008;62(Suppl 2):ii87–95.
10. National Committee for Clinical Laboratory Standards. Performance standards for antimicrobial susceptibility testing: eighth informational supplement (M100-S8). Wayne, PA: NCCLS, 1998:18.
11. Centers for Disease Control and Prevention. Effects of new penicillin susceptibility breakpoints for Streptococcus pneumoniae - United States, 2006–2007. MMWRMorb Mortal Wkly Rep 2008;57:1353–5. Available at: http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5750a2.htm#tab (accessed 15 September 2009)
12. World Health Organization. Acute respiratory infections in children. WHO Available at http://www.who.int/fch/depts/cah/resp_infections/en/ (accessed 15 September 2009).
13. British Thoracic Society. BTS guidelines for the management of community acquired pneumonia in childhood. Thorax 2002;57(Suppl 1):i1–24. (accessed 15 September 2009).
14. Tan TQ, Mason EOJr, Wald ER, et al . Clinical characteristics of children with complicated pneumonia caused by Streptococcus pneumoniae. Pediatrics 2002;110:1–6.
15. Menendez R, Torres A . Treatment failure in community-acquired pneumonia. Chest 2007;132:1348–55.
16. Aspa J, Rajas O, Rodriguez de Castro F, et al . Drug-resistant pneumococcal pneumonia: clinical relevance and related factors. Clin Infect Dis 2004;38:787–98.
17. Pallares R, Gudiol F, Linares J, et al . Risk factors and response to antibiotic therapy in adults with bacteremic pneumonia caused by penicillin-resistant pneumococci. N Engl J Med 1987;317:18–22.
18. Giachetto G, Pirez MC, Nanni L, et al . Ampicillin and penicillin concentration in serum and pleural fluid of hospitalized children with community-acquired pneumonia. Pediatr Infect Dis J 2004;23:625–9.