Skip Navigation

An International Prospective Study of Pneumococcal Bacteremia: Correlation with In Vitro Resistance, Antibiotics Administered, and Clinical Outcome

  1. Victor L. Yu1,
  2. Christine C. C. Chiou5,
  3. Charles Feldman8,
  4. Ake Ortqvist10,
  5. Jordi Rello11,
  6. Arthur J. Morris15,
  7. Larry M. Baddour2,
  8. Carlos M. Luna13,
  9. David R. Snydman3,
  10. Margaret Ip7,
  11. Wen Chien Ko6,
  12. M. Bernadete F. Chedid14,
  13. Antoine Andremont12,
  14. Keith P. Klugman4,9, and
  15. International Pneumococcal Study Group
  1. 1Division of Infectious Disease, University of Pittsburgh, Pennsylvania
  2. 2Mayo Clinic, Rochester, Minnesota
  3. 3Division of Geographic Medicine and Infectious Diseases, Tufts-New England Medical Center, Tufts University School of Medicine, Boston, Massachusetts
  4. 4Department of International Health, Rollins School of Public Health, Emory University, Atlanta, Georgia
  5. 5Department of Pediatrics, Veterans General Hospital-Kaohsiung, National Yang Ming University, Taipei
  6. 6Division of Infectious Diseases, National Cheng Kung University Hospital, Tainan, Taiwan
  7. 7Department of Microbiology, Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, Hong Kong
  8. 8Pulmonology Division, Department of Medicine, Johannesburg Hospital and University of the Witwaterstrand, Johannesburg, South Africa
  9. 9Wits/Medical Research Council/National Health Laboratory Service Respiratory and Meningeal Pathogens Research Unit, Johannesburg, South Africa
  10. 10Unit of Infectious Diseases, Department of Communicable Diseases and Prevention, Karolinska Institute, Karolinska Hospital, Department of Communicable Diseases and Prevention, Stockholm, Sweden
  11. 11University Hospital Joan XXIII, University Rovira and Virgili, Tarragona, Spain
  12. 12Laboratorie de Bacteriologie, Hopital Bichat-Claude Bernard AP-HP, Paris, France
  13. 13Division of Pulmonary Medicine, Hospital de Clinicas, Universidad de Buenos Aires, Buenos Aires, Argentina
  14. 14Postgraduate Program in Pulmonology, Federal University of Rio Grande do Sol, Hospital de Clinicas de Porto Alegre, Brazil
  15. 15Clinical Microbiology Laboratory, Auckland Hospital, Auckland, New Zealand
  1. Reprints or correspondence: Dr. Victor L. Yu, Infectious Disease Section, VA Medical Center, University Drive C, Pittsburgh, PA 15240 (vly+{at}pitt.edu).
 
Next Section

Abstract

We performed a prospective, international, observational study of 844 hospitalized patients with blood cultures positive for Streptococcus pneumoniae. Fifteen percent of isolates had in vitro intermediate susceptibility to penicillin (minimum inhibitory concentration [MIC], 0.12–1 μg/mL), and 9.6% of isolates were resistant (MIC, ⩾2 μg/mL). Age, severity of illness, and underlying disease with immunosuppression were significantly associated with mortality; penicillin resistance was not a risk factor for mortality. The impact of concordant antibiotic therapy (i.e., receipt of a single antibiotic with in vitro activity against S. pneumoniae) versus discordant therapy (inactive in vitro) on mortality was assessed at 14 days. Discordant therapy with penicillins, cefotaxime, and ceftriaxone (but not cefuroxime) did not result in a higher mortality rate. Similarly, time required for defervescence and frequency of suppurative complications were not associated with concordance of β-lactam antibiotic therapy. β-Lactam antibiotics should still be useful for treatment of pneumococcal infections that do not involve cerebrospinal fluid, regardless of in vitro susceptibility, as determined by current NCCLS breakpoints.

The alarming increase in the prevalence of penicillin resistance among Streptococcus pneumoniae has been well publicized, and, as a result, shifts in the empirical antibiotic therapy administered for community-acquired pneumonia are occurring. Considerable surveillance data have been reported for the in vitro susceptibility of invasive and noninvasive S. pneumoniae isolates [1–5]. However, the clinical relevance is uncertain, because there is a paucity of data on the outcome of patients infected with penicillin-nonsusceptible pneumococci [6].

Some studies have suggested that antibiotic resistance in S. pneumoniae is not clinically relevant [7–19], whereas others have reported higher mortality rates among patients infected with penicillin-nonsusceptible S. pneumoniae [20–22]. However, limitations exist in all of these studies, including retrospective study design [12, 14, 17, 20, 22], lack of clinical data [20], small sample size [13, 14, 18], inclusion of nonbacteremic patients [11, 13, 14, 18, 19], and failure to stratify for severity of illness [21]. These limitations have precluded an evidence-based solution to this important clinical question.

Therefore, we conducted a prospective, international, large-scale, observational study involving patients with pneumococcal bacteremia. We evaluated risk factors for emergence of drug resistance, performed in vitro susceptibility testing for blood isolates, and, most importantly, recorded data on the antibiotics administered, including timing of antibiotic therapy in relation to the onset of symptoms, dose, duration of therapy, and route of administration. Our primary objective was to assess the clinical impact of emerging drug resistance for S. pneumoniae to clarify antibiotic recommendations for community-acquired pneumonia.

Previous SectionNext Section

Materials And Methods

Study design. From December 1998 through January 2001, 844 consecutive hospitalized adults with pneumococcal bacteremia were enrolled in 21 hospitals in 10 countries (South Africa, 29.6% of patients; United States, 13.6%; Sweden, 12.1%; Spain, 11.7%; New Zealand, 11.7%; Taiwan, 5.7%; Argentina, 5.2%; Brazil, 3.9%; Hong Kong, 3.9%; and France, 2.6%). The health care centers in the United States were located in Boston; Knoxville, Tennessee; Pittsburgh; and Washington, D.C.

The collaborative study institutions and investigators were not selected at random. Investigators who had previously conducted clinical studies on pneumococcal pneumonia and investigators who were located at institutions in countries or cities reporting a notable prevalence of drug-nonsusceptible pneumococci were recruited. Furthermore, they had to be sufficiently motivated to participate in a large-scale clinical study without formal funding. Institutional review board approval was obtained in accordance with local requirements.

Patients were observed for ⩾14 days after the onset of bacteremia to assess clinical outcome, including mortality and complications of infection. The study was observational in that administration of antimicrobial agents and other therapies was managed by the patients' physicians and not by the investigators. The end point was mortality at 14 days after the first blood culture positive for S. pneumoniae was obtained.

In vitro susceptibility testing and serotyping. A total of 793 (94%) of 844 blood isolates were available for in vitro susceptibility testing and serotyping. All isolates were frozen at -70°C and then sent on Dorset egg medium to the Veterans Affairs Pittsburgh Special Pathogens Laboratory, where the identity of S. pneumoniae was confirmed for each isolate.

The susceptibility results for all antimicrobials administered for pneumococcal infections were determined using Etest (AB Biodisk), in accordance with the manufacturer's instructions, and on the basis of NCCLS breakpoints for pneumonia [23]. The term “nonsusceptible” was used when the MIC of penicillin for S. pneumoniae was intermediate (MIC, 0.12–1.0 μg/mL) or indicative of resistance (MIC, ⩾2 μg/mL). High-level penicillin resistance was defined as an MIC of ⩾3 μg/mL, as determined by Etest. Although Etest results are approved by the US Food and Drug Administration for susceptibility testing of S. pneumoniae, persons at 2 independent laboratories (M. Jacobs and R. N. Jones) retested all penicillin-nonsusceptible isolates in a blinded protocol using NCCLS broth microdilution methods. The concordance rates between Etest and broth microdilution were 100% in one laboratory and 81% in the other.

Definitions. “Underlying chronic disease” was defined as presence of heart disease, liver disease, renal disease, lung disease, or diabetes mellitus. “Underlying disease or risk factor associated with immunosuppression” was defined as HIV infection, receipt of a splenectomy, hematological malignancy, autoimmune disorder, receipt of a transplant, and receipt of cancer chemotherapy within 4 weeks before onset of bacteremia. “Prior antibiotic therapy” was defined as use of antibiotics for ⩾1 day within 3 months before the onset of bacteremia.

Antibiotic therapy. “Concordant therapy” was defined as receipt for the first 2 days after the blood was obtained for culture of a single antibiotic that had in vitro activity (i.e., neither intermediate nor resistant in vitro) against the S. pneumoniae isolated. “Discordant therapy” was defined as receipt for the first 2 days after the blood sample was obtained for culture of a single antibiotic that was inactive in vitro against the S. pneumoniae isolated.

Severity of illness was assessed using the Pitt bacteremia [24] and APACHE II scores at the time of the positive blood culture result. The Pitt bacteremia score was calculated within 1 day after hospital admission for all patients on the basis of vital signs and mental status. “Critical illness” was defined as a Pitt bacteremia score of >4 points; this score has been validated in 10 observational studies of bloodstream infection. The APACHE II score was applied only for patients admitted to the intensive care unit (ICU) at the time of the positive blood culture result. Nosocomial bacteremia was considered to have occurred if the first positive blood culture result was obtained >3 days after hospital admission.

Statistical analysis. Clinical and laboratory data for analysis were entered into a computerized database (Prophet Systems, National Institute of Health; Bethesda, MD). Categorical data were analyzed using the χ2 or Fisher's exact test. Continuous variables were compared using the t test or the Mann-Whitney U test. A logistic regression model was used to examine the effects of multiple risk factors on penicillin resistance and mortality. The factors in the regression model included those found to be significant by univariate analysis and those previously hypothesized to affect penicillin resistance or mortality. Statistical significance was defined as P < .05. In this report, P ⩾ .05 was considered not significant (NS). If the P value was .05–.20, it was still considered NS, but the precise P value is given in parentheses (see tables 1–3).

Patient characteristics. Eight hundred forty-four consecutive patients with pneumococcal bacteremia were enrolled during the study period. The mean patient age was 52.1 years (median, 51 years; range, 15–97 years). A total of 48.2% of patients had ⩾1 chronic underlying disease, and 42.1% had an underlying disease or risk factor associated with immunosuppression; 32.4% of patients had >1 underlying disease. A total of 19.3% of patients were critically ill, as defined in Methods. A total of 20.2% of patients were admitted to ICU with a mean APACHE II score of 18.4 (range, 3–45). The Pitt bacteremia and APACHE II scores were highly correlated with each other (P = .0001); the predictive value for mortality was superior for the Pitt bacteremia score, compared with the APACHE II score (81.0 vs. 74.9).

Microbiology results. Seven hundred ninety-three pneumococcal isolates were available for in vitro susceptibility testing. Fifty-one (6%) of the isolates did not survive initial storage at -70°C. Of interest, all isolates survived handling and shipping to the central reference laboratory. To exclude the possibility of bias, the following clinical factors were compared for the patients infected with the 51 missing isolates versus the other patients: ICU admission, immunosuppression, community-acquired versus nosocomially acquired bacteremia, suppurative complications, and mortality; no significant differences were found. Moreover, the antibiotic susceptibility patterns of the 51 missing isolates, as initially determined at the original hospital, were not significantly different from the susceptibility patterns of the 793 isolates, as determined at the reference laboratory.

One hundred nineteen (15%) of the 793 pneumococcal isolates had intermediate susceptibility to penicillin, and 76 (9.6%) were resistant. Thirteen pneumococci had an MIC of penicillin of ⩾3 μg/mL and were considered to be highly resistant. The prevalence for penicillin-nonsusceptible pneumococci, by region, was as follows: Taiwan, 57.4% of isolates; Hong Kong, 53.1%; United States, 30.8%; France, 29.4%; Spain, 27.6%; South Africa, 24.0%; Brazil, 18.7%; New Zealand, 18.6%; Sweden, 6.9%; Argentina, 2.6%.

Prior antibiotics. A total of 20.3% of patients received ⩾1 antibiotic for 1–70 days (median duration, 8 days) ⩽3 months before this episode of pneumococcal bacteremia. A total of 10.6% received penicillins (i.e., penicillin, ampicillin, amoxicillin, amoxicillin-clavulanate, or ampicillin-sulbactam), 13.4% received cephalosporins, 2.1% received macrolides, 3.8% received quinolones, and 4.3% received other antibiotics (i.e., clindamycin, chloramphenicol, tetracycline, trimethoprim-sulfamethoxazole, rifampin, or vancomycin). Note that many patients received >1 antibiotic.

Antibiotic therapy. Three hundred eighty-nine patients received only 1 antibiotic for at least the first 2 days after blood samples were obtained for culture. Of the 389 episodes of pneumococcal bacteremia, 21 pneumococcal isolates were not available, and 8 involved patients who received antibiotics for which NCCLS breakpoints for pneumococci were not available. Thus, data for 360 patients were evaluable for adequacy of antibiotic therapy. Three hundred thirty-five (93.1%) of 360 patients received concordant therapy, and 25 (6.9%) received discordant therapy.

Risk factors for penicillin nonsusceptibility. Statistically significant risk factors for penicillin resistance on univariate analysis included liver disease, cancer, receipt of corticosteroid therapy, receipt of a transplant, chronic underlying disease, underlying disease or risk factor associated with immunosuppression, and nosocomially acquired bacteremia (data not shown). On multivariate analysis (table 1), only 2 factors were found to be statistically significant: underlying diseases or risk factors associated with immunosuppression (P = .0001) and prior receipt of antibiotics (P = .009).

Outcome. The mortality rate for patients with pneumococcal bacteremia was 16.9% (139 of 823 patients died). A total of 64.5% of the deaths occurred within the first 3 days after a blood sample was obtained for culture (figure 1). The following factors were significantly associated with mortality on univariate analyses: prior hospitalization, chronic underlying disease, underlying disease or risk factor associated with immunosuppression (especially HIV infection), bilateral infiltrates on chest radiograph obtained at hospital admission, meningitis, suppurative complications, and old age. Fever, rigors, and chest pain were each significantly associated with survival, as has been reported elsewhere [25]. Although in vitro penicillin nonsusceptibility was associated with mortality on univariate analysis, statistical significance disappeared on the multivariate analysis. Age of >65 years (P = .0001), severity of illness (P = .0001), and presence of an underlying disease or risk factor associated with immunosuppression (P = .0001) remained significant on multivariate analysis (table 2).

Figure 1
View larger version:
Figure 1

Distribution of deaths in relation to the time initial blood cultures were obtained. A total of 64.5% of deaths occurred within 3 days after a blood culture positive for Streptococcus pneumoniae was obtained.

Concordant and discordant antibiotic therapy. To assess the impact of antimicrobial susceptibility on patient outcome, we examined the concordance of antimicrobial therapy within the first 2 days after the blood sample was obtained for culture. This analysis was confined to the 360 patients receiving monotherapy. Receipt of discordant therapy did not result in a statistically significantly higher mortality rate. The mortality rate was high for critically ill patients in both concordant and discordant therapy groups. Patients who were infected with pneumococci that were not susceptible to cefuroxime (median MIC by Etest, 3 μg/mL; median MIC by broth microdilution assay, 8 μg/mL) but who were treated with cefuroxime experienced a significantly higher mortality rate (4 [5.8%] of 53 patients died) with cefuroxime concordance, compared with cefuroxime discordance (4 [36.4%] of 11; P = .02). On the other hand, discordant therapy involving penicillins, cefotaxime, and ceftriaxone did not result in a significantly higher mortality rate, regardless of severity of illness (table 3). We also examined the effect of therapy concordance on mortality for the total sample, with the following exclusions: (1) patients with meningitis were excluded, and (2) patients who died ⩽3 days after blood cultures were performed were excluded. The above 2 groups were also stratified by severity of illness. The results for these subgroups were similar to the results for all patients; only discordant therapy involving cefuroxime was significantly associated with a higher mortality rate (P = .01; data not shown; none of the patients had meningitis). This association was most pronounced in patients who were not critically ill.

Clinical response. Time to defervescence was compared for 3 patient groups: (1) patients infected with penicillin-susceptible pneumococci versus those infected with penicillin-nonsusceptible pneumococci (figure 2A), (2) patients receiving concordant therapy versus those receiving discordant therapy (figure 2B), and (3) patients receiving concordant therapy involving cefuroxime only versus those receiving discordant therapy involving cefuroxime only (figure 2C). Clinical response, as measured by time to defervescence in days, was also unaffected by the in vitro susceptibility of the infecting pneumococci: 19.2% of patients infected with in vitro nonsusceptible pneumococci had defervescence on day 5, compared with 12.5% of patients infected with susceptible pneumococci (P = NS; figure 2A). Time to defervescence also did not significantly differ on the basis of concordant versus discordant therapy: 15% of recipients of concordant therapy had defervescence by day 5, and 24% in the discordant therapy group had defervescence by day 5 (P = NS; figure 2B). Of interest, discordant therapy involving cefuroxime showed a trend for slower defervescence, although statistical significance was not attained (P = .09; figure 2C).

Figure 2
View larger version:
Figure 2

Clinical response (time to defervescence) among patients with pneumococcal bacteremia. A, Penicillin-susceptible vs. -nonsusceptible pneumococci. No statistically significant difference in clinical response (duration of fever) was seen among patients infected with penicillin-susceptible vs. -nonsusceptible pneumococci (P = NS, by Mantel-Cox test). B, Concordant vs. discordant therapy. No significant difference in the duration of fever was seen in patients receiving concordant vs. discordant therapy. (P = NS, by Mantel-Cox test). C, Discordant therapy with cefuroxime. Discordant therapy with cefuroxime resulted in prolongation of fever, but statistical significance was not achieved (P = .09 [NS], by Mantel-Cox text). NS, not significant.

Suppurative complications of meningitis (6.9%), empyema (3.3%), endocarditis (0.6%), lung abscess (0.5%), and pericarditis (0.4%) did not occur significantly more frequently in patients infected with penicillin-nonsusceptible pneumococci, compared with those infected with penicillin-susceptible pneumococci (13% vs. 10.5%, respectively; P = NS). The frequency of suppurative complications was somewhat higher among patients receiving discordant therapy than among patients receiving concordant therapy, but statistical significance was not attained (17% vs 8.9%; P = .13).

Previous SectionNext Section

Discussion

Although emerging antibiotic resistance among pneumococci is a well-publicized concern, the impact of penicillin resistance on the outcome of pneumococcal pneumonia has been controversial. In contrast to previous studies that have assessed this issue, our study was prospective, and specific details were recorded for the antibiotics administered, including dose, route of administration, and duration of therapy, as well as the clinical response to these antibiotics. Bacteremia was required for enrollment because it is an impeccable “gold standard”; the sputum isolates reported in large-scale surveillance studies without accompanying clinical data may not be pathogenic. Furthermore, patients can be colonized with >1 strain of pneumococcus in the upper respiratory tract, which could result in difficulties in correctly identifying the infecting strain, particularly for patients who have received antibiotics [26]. Bacteremia is also an established risk factor for death due to pneumonia, and, therefore, it involves the subgroup of patients most likely to die. This would maximize the clinical relevance of our study with respect to patient treatment.

One hundred ninety-five (24.6%) of the 793 pneumococcal isolates were not susceptible to penicillin (MIC for intermediate resistance, 0.12–1 μg/mL; MIC for resistance, ⩾2 μg/mL), as determined by NCCLS criteria [23]. In our study, numerous factors were significantly associated with nonsusceptibility on univariate analysis (see Results), but only underlying disease or risk factors associated with immunosuppression and prior antibiotic use were significantly associated with penicillin nonsusceptibility on multivariate analyses (table 1). Prior antibiotic use as a risk factor for penicillin nonsusceptibility has been documented in numerous studies [9, 22, 27–29].

Table 1
View larger version:
Table 1

Multivariate analysis of factors potentially associated with penicillin resistance in vitro.

The mortality rate at 14 days was 16.9%, which is similar to the rate reported in other studies of pneumococcal bacteremia [18, 30]. Age >65 years, severity of illness, and underlying disease or risk factors involving immunosuppression were significantly associated with mortality on multivariate analysis (table 2).

Table 2
View larger version:
Table 2

Multivariate analysis of factors potentially associated with mortality in 844 patients.

It is difficult to assess the effect of antibiotics when multiple antibiotics are given with initial therapy [11, 31]. In our large-scale study, 389 patients received monotherapy, and data for 360 of these patients were evaluable. Within this subgroup, patients who received discordant therapy did not have a higher mortality rate than did patients who received concordant therapy, for all patients and for the subgroup of patients receiving penicillins (penicillin, ampicillin, and amoxicillin-clavulanate) and third-generation cephalosporins. However, patients who were treated with cefuroxime, a second-generation cephalosporin, and who were infected with pneumococci with in vitro nonsusceptibility to cefuroxime experienced significantly higher mortality rates on both univariate and multivariate analyses (table 3). Of note, after we stratified groups by severity of illness, the effect of discordant therapy with cefuroxime became even more pronounced in patients who were not critically ill (P = .01). On the other hand, discordant therapy involving cefotaxime or ceftriaxone, 2 third-generation cephalosporins with higher intrinsic activity, did not result in higher mortality rates, regardless of severity of illness (table 3).

Table 3
View larger version:
Table 3

Factors evaluated for mortality in 360 patients receiving antibiotic monotherapy on multivariate analyses.

The severity of illness was consistently the most important determinant of impending death in all analyses. The mortality rate was high among critically ill patients, irrespective of the antibiotics administered or receipt of concordant and discordant therapy. This may be explained by the 1964 study by Austrian and Gold [32] in which mortality associated with pneumococcal bacteremia occurred mainly ⩽3 days after positive blood cultures were obtained, as is also shown in figure 1 of our study. In this early phase, excessive and dysregulated release of cytokines, nitric oxide, and hypotensive molecules, such as kinins, leads to septic shock. Antibiotics appear to have little effect in minimizing mortality. Critical illness simply overwhelmed the salutary effects of active in vitro antibiotic therapy such that the issue of concordant versus discordant therapy became irrelevant.

By excluding patients who died during the first 4 days of hospitalization (as in the study by Austrian and Gold [32]), investigators from the Centers for Disease Control and Prevention found that patients infected with pneumococci that had high-level penicillin resistance experienced significantly higher mortality rates [19]. Similarly, when we excluded patients who were critically ill, the association between mortality and the use of discordant therapy with cefuroxime became statistically significant; on the other hand, discordant therapy involving penicillins, cefotaxime, or ceftriaxone had no correlation with mortality in this subgroup.

The pharmacodynamic features of individual antibiotics may explain, in part, our observations. The effectiveness of β-lactam antibiotics has been correlated with the duration of serum or tissue antibiotic concentrations above the MIC for the infecting organisms [33]. Eleven patients received monotherapy with cefuroxime for at least the first 48 h of therapy. Four of 11 patients infected with cefuroxime-resistant pneumococci died. Eight of the 11 patients and all 4 patients who died received the standard dose of cefuroxime (750 mg iv q8h). Pharmacodynamic studies suggest that cephalosporin concentrations should exceed the MIC for ⩾50% of the dosing interval to ensure bacterial eradication and clinical cure in respiratory tract infections [26]. An analysis of the pharmacodynamics of cefuroxime at 750 mg 3 times per day reveals that the drug concentration exceeds 2 μg/mL for 50% of the dosing interval. Coverage of an MIC of 4 μg/mL for 50% of the dosing interval requires dosing at 1500 mg intravenously every 8 h [33–35]. Thus, our data are consistent with the concept that the higher dose should be recommended where cefuroxime-resistant pneumococci are likely to be encountered. In contrast, standard parenteral doses of ampicillin, penicillins, cefotaxime, and ceftriaxone achieve concentrations greater than the MIC for 75%–100% of the dosing interval for both susceptible and nonsusceptible pneumococci. It is noteworthy that bacteriological failure of first-generation cephalosporins and cefuroxime in patients with pneumococcal infection has been reported elsewhere [36–38].

One limitation of our study is that it could be argued that the low incidence of penicillin nonsusceptibility seen in this study may not be representative of the overall population. However, study hospitals in countries reporting high-level resistance (France, Spain, Hong Kong, and Taiwan) were specifically recruited for participation. Tennessee also has a notably higher frequency of penicillin-nonsusceptible pneumococci than do other states in the United States and was thus represented in this study [4, 39]. If there is a bias of study hospitals toward in vitro susceptibility of pneumococci, it probably is toward a greater prevalence of resistance.

A second limitation is that this was an observational study, and the type of antibiotic administered could not be controlled. However, no controlled interventional trial has even been performed to assess the clinical significance of in vitro pneumococcal resistance.

A third limitation is that the macrolides and quinolones were not widely used as monotherapy. Thus, we could not assess the impact of in vitro resistance on mortality for these 2 classes of antibiotics. Of interest, both macrolides (33%) and quinolones (10%) were commonly used as components of combination antibiotic therapy.

In summary, with the exception of cefuroxime, the current NCCLS breakpoints for S. pneumoniae were not predictive of clinical outcome. Although in vitro resistance is increasing among pneumococci, it appears that this increase has not yet become clinically relevant for treatment of pneumococcal pneumonia.

S. pneumoniae is not only the most common cause of community-acquired pneumonia but the one that most frequently causes death. Its widely publicized increasing in vitro drug resistance has led to shifts in antibiotic prescription by physicians for community-acquired pneumonia [40–42]. Indeed, a premature shift to classes of antibiotics with increased potency against drug-resistant S. pneumoniae, such as quinolones, may render these valuable antibiotics less useful in the future. Quinolone-resistant pneumococci have already been reported as use of these antibiotics has increased [1, 43, 44]. Our study confirms that β-lactam agents, including penicillin, are still effective for the treatment of penicillin-nonsusceptible pneumococcal pneumonia. In our study, penicillin was found to still be in wide use in Sweden, South Africa, and New Zealand. It may be argued that penicillin should be more widely used in the United States and elsewhere.

Previous SectionNext Section

The International Pneumococcal Study Group

Argentina. Hospital de Clinicas, Buenos Aires: Carlos M. Luna (principal investigator), Ricardo Mosquera, Carmen de Mier, Angela Famiglietti, and Carlos Vay. Hospital Santmarina, Tandil: Jorge Gentile and Monica Sparo.

Brazil. Hospital de Clinicas de Porto Alegre: Maria Bernadete F. Chedid (principal investigator), Serigo Menna Barreto, and Afonso L. Barth. Santa Casa de Misericordia de Porto Alegre: Cicero Dias. Hospital Nossa Senhora da Conceicao: Breno Riegel dos Santos.

France. Hospital Bichat-Claude Bernard, Paris: Antoine Andremont (principal investigator), Karine Grenet, and Hyam Maunieme.

Hong Kong. Chinese University of Hong Kong: David Hui (principal investigator), Margaret Ip, and Donald Lyon.

New Zealand. Auckland Hospital, Auckland: Arthur J. Morris (principal investigator), Sally A. Roberts, Dragana Drinkovic, and Susan L. Taylor.

South Africa. Johannesburg Hospital, Johannesburg: Charles Feldman (principal investigator), Keith P. Klugman (co-principal investigator), Anne von Gottberg, Xoliswa Poswa, and Rajen Morar.

Spain. University Hospital Joan XXIII, Tarragona: Jordi Rello (principal investigator), Emili Diaz, and Jose M. Santamaria; Corporacio Sanitaria Parc Tauli, Sabadell: Dolous Mariscal, Dionisa Fontanals, M. Lujan, and Miquel Gallego.

Sweden. Karolinska Hospital, Stockholm: Ake Ortqvist (principal investigator) and Margareta Rylander.

Taiwan. Tri-Service General Hospital, Taipei: Feng-Yee Chang (principal investigator). National Cheng Kung University Hospital, Tainan: Wen-Chien Ko (principal investigator). St. Martin De Porres Hospital: Wen-Pin Wu and Chia Yi.

United States. New England Medical Center, Boston: David R. Snydman (principal investigator), Laurie Barefoot, and Laura McDermott. National Naval Medical Center, Bethesda, Maryland: David L. Blazes and Gregory Marin. University of Tennessee Medical Center, Knoxville: Larry M. Baddour (principal investigator), Madana Mobasseri. University of Pittsburgh and Veterans Affairs Medical Center, Pittsburgh: Victor L. Yu (principal investigator), Dong Hoon Daniel Kim, and John D. Rihs.

Previous SectionNext Section

Acknowledgments

We thank Michael Jacobs and Ronald Jones for reconfirmation of Etest results by microtiter broth dilution. We thank Michael Jacobs and Ronald Dagen for their extensive and perceptive analyses. The following individuals also provided valuable critiques: Lionel Mandell, Michael Niederman, Sheldon Kaplan, Donald Low, James T. Jorgensen, Joshua Metlay, and Ian Friedland.

Previous SectionNext Section

Footnotes

  • Financial support: After the study protocol was approved and initiated by the investigators, unrestricted educational grant support was obtained for in vitro susceptibility testing and laboratory costs from Wyeth-Ayerst Laboratories, Glaxo Smith-Kline Beecham, Roche Labs, Pfizer Laboratories, and Abbott Laboratories.

  • Received December 10, 2002.
  • Accepted June 12, 2003.
Previous Section
 

References

  1. ↵
    1. Chen DK,
    2. McGeer A,
    3. de Azavedo JC,
    4. Low DE
    . Decreased susceptibility of Streptococcus pneumoniae to fluoroquinolones in Canada. N Engl J Med 1999;341:233-9.
    CrossRefMedlineWeb of Science
  2. ↵
    1. Chiou CC,
    2. Liu YC,
    3. Hwang TS,
    4. et al
    . Extremely high prevalence of nasopharyngeal carriage of penicillin-resistant Streptococcus pneumoniae among children in Kaohsiung, Taiwan. J Clin Microb 1998;36:1933-7.
    Abstract/FREE Full Text
    1. Song J,
    2. Lee NY,
    3. Ichiyama S,
    4. et al
    . Spread of drug-resistant Streptococcus pneumoniae in Asian countries: Asian Network for Surveillance of Resistant Pathogens (ANSORP) Study. Clin Infect Dis 1999;28:1206-11.
    Abstract/FREE Full Text
  4. ↵
    1. Whitney CG,
    2. Farley MM,
    3. Hadler J,
    4. et al
    . Increasing prevalence of multidrug-resistant Streptococcus pneumoniae in the United States. Active Bacterial Core Surveillance Program of the Emerging Infections Program Network. N Engl J Med 2000;343:1917-24.
    CrossRefMedlineWeb of Science
  5. ↵
    1. Doern GV,
    2. Heilmann KP,
    3. Huynh HK,
    4. Rhomberg PR,
    5. Coffman SL,
    6. Brueggemann AB
    . Antimicrobial resistance among clinical isolates of Streptococcus pneumoniae in the United States during 1999–2000, including a comparison of resistance rates since 1994–1995. Antimicrob Agents Chemother 2001;45:1721-9.
    Abstract/FREE Full Text
  6. ↵
    1. Bishai W
    . The in-vivo and in-vitro paradox in pneumococcal respiratory tract infections. J Antimicrob Chemother 2002;49:433-6.
    FREE Full Text
  7. ↵
    1. Siegel RE
    . The significance of serum vs tissue levels of antibiotics in the treatment of penicillin-resistant Streptococcus pneumoniae and community-acquired pneumonia. Chest 1999;116:535-8.
    FREE Full Text
    1. Amsden GW,
    2. Amankwa K
    . Pneumococcal resistance: the treatment challenge. Ann Pharmacother 2001;35:480-7.
    Abstract
  9. ↵
    1. Deeks S,
    2. Palacio R,
    3. Ruvinsky R,
    4. et al
    . Risk factors and course of illness among children with invasive penicillin-resistant Streptococcus pneumoniae. Pediatrics 1999;103:409-13.
    Abstract/FREE Full Text
    1. Friedland IR,
    2. Klugman KP
    . Antibiotic-resistant pneumococcal disease in South African children. Am J Dis Child 1992;146:920-3.
    Abstract/FREE Full Text
  11. ↵
    1. Pallares R,
    2. Linares J,
    3. Vadillo M
    . Resistance to penicillin and cephalosporin and mortality from severe pneumococcal pneumonia in Barcelona, Spain. N Engl J Med 1995;333:474-80.
    CrossRefMedlineWeb of Science
  12. ↵
    1. Castillo EM,
    2. Rickman LS,
    3. Brodine SK,
    4. et al
    . Streptococcus pneumoniae bacteremia in an era of penicillin resistance. Am J Infect Control 2000;28:239-43.
    CrossRefMedlineWeb of Science
  13. ↵
    1. Watanabe H,
    2. Sato S,
    3. Kawakami K,
    4. et al
    . A comparative clinical study of pneumonia by penicillin-resistant and -sensitive Streptococcus pneumoniae in a community hospital. Respirology 2000;5:59-64.
    CrossRefMedline
  14. ↵
    1. Wu TT,
    2. Hsueh PR,
    3. Lee LN,
    4. et al
    . Pneumonia caused by penicillin-nonsusceptible Streptococcus pneumoniae clinical characteristics, prognostic factors and outcomes. J Formos Med Assoc 2000;99:18-22.
    MedlineWeb of Science
    1. Henriques B,
    2. Kalin M,
    3. Ortqvist A,
    4. et al
    . Molecular epidemiology of Streptococcus pneumoniae causing invasive disease in 5 countries. J Infect Dis 2000;182:833-9.
    Abstract/FREE Full Text
    1. Gomez-Barreto D,
    2. Caldren-Jaimes E,
    3. Rodriguez RS,
    4. de los Monteros LE
    . Clinical outcome of invasive infections in children caused by highly penicillin-resistant Streptococcus pneumoniae compared with infections caused by penicillin-susceptible strains. Arch Med Res 2000;31:592-8.
    CrossRefMedlineWeb of Science
  17. ↵
    1. Kim BN,
    2. Bae IG,
    3. Kim MN,
    4. Park SJU,
    5. Wood JH,
    6. Kim YS
    . Risk factors for penicillin resistance and mortality in Korean adults with Streptococcus pneumoniae bacteremia. Eur J Clin Microbiol Infect Dis 2002;21:35-42.
    CrossRefMedlineWeb of Science
  18. ↵
    1. Ewig S,
    2. Ruiz M,
    3. Torres A,
    4. et al
    . Pneumonia acquired in the community through drug-resistant Streptococcus pneumoniae. Am J Respir Crit Care Med 1999;159:1835-42.
    Abstract/FREE Full Text
  19. ↵
    1. Pallares R,
    2. Capdevila O,
    3. Linares J,
    4. et al
    . The effect of cephalosporin resistance on mortality in adult patients with nonmeningeal systemic pneumococcal infections. Am J Med 2002;113:120-6.
    CrossRefMedlineWeb of Science
  20. ↵
    1. Metlay JP,
    2. Hofmann J,
    3. Cetron MS,
    4. et al
    . Impact of penicillin susceptibility on medical outcomes for adult patients with bacteremic pneumococcal pneumonia. Clin Infect Dis 2000;30:520-8.
    Abstract/FREE Full Text
  21. ↵
    1. Feikin DR,
    2. Schuchat A,
    3. Kolczak M,
    4. et al
    . Mortality from invasive pneumococcal pneumonia in the era of antibiotic resistance, 1995–1997. Am J Public Health 2000;90:223-9.
    Abstract/FREE Full Text
  22. ↵
    1. Turett GS,
    2. Blum S,
    3. Fazal BA,
    4. Justman JE,
    5. Telzak EE
    . Penicillin resistance and other predictors of mortality in pneumococcal bacteremia in a population with high human immunodeficiency virus seroprevalence. Clin Infect Dis 1999;29:321-7.
    MedlineWeb of Science
  23. ↵
    1. NCCLS
    . NCCLS document M100-S12. Wayne, PA: NCCLS; 2002. Performance standards for antimicrobial testing. 12th informational supplement.
    1. Chow JW,
    2. Yu VL
    . Combination antibiotic versus monotherapy for gram-negative bacteremia: a commentary. Int J Antimicrob Agents 1999;11:7-12.
    CrossRefMedlineWeb of Science
  25. ↵
    1. Ortqvist A,
    2. Hedlund J,
    3. Grillner L,
    4. et al
    . Aetiology, outcome, prognostic factors in community-acquired pneumonia requiring hospitalization. Eur Respir J 1990;3:1105-13.
    Abstract
  26. ↵
    1. Dagan R,
    2. Klugman KP,
    3. Craig WA,
    4. Baquero R
    . Evidence to support the rationale that bacterial eradication in respiratory tract infection is an important aim of antimicrobial therapy. J Antimicrob Chemother 2001;47:129-40.
    Abstract/FREE Full Text
  27. ↵
    1. Nava JM,
    2. Garau J,
    3. Lite J,
    4. et al
    . Predictive factors for invasive disease due to penicillin-resistant Streptococcus pneumoniae a population-based study. Clin Infect Dis 1994;19:884-90.
    Abstract/FREE Full Text
    1. Bedos J-P,
    2. Chevret S,
    3. Chastang C,
    4. Geslin P,
    5. Regnier B
    . Epidemiological features of and risk factors for infection by Streptococcus pneumoniae strains with diminished susceptibility to penicillin: findings of a French survey. French Cooperative Pneumococcus Study Group. Clin Infect Dis 1996;22:63-72.
    Abstract/FREE Full Text
  29. ↵
    1. Clavo-Sanchez AJ,
    2. Giron-Gonzalez JA,
    3. Lopez-Prieto D,
    4. et al
    . Multivariate analysis of risk factors for infection due to penicillin-resistant and multidrug-resistant Streptococcus pneumoniae a multicenter study. Clin Infect Dis 1997;24:1052-9.
    Abstract/FREE Full Text
  30. ↵
    1. Kalin M,
    2. Ortqvist A,
    3. Almela M,
    4. et al
    . Prospective study of prognostic factors in community-acquired bacteremic pneumococcal disease in 5 countries. J Infect Dis 2000;182:840-7.
    Abstract/FREE Full Text
  31. ↵
    1. Moroney JF,
    2. Fiore AE,
    3. Harrison LH,
    4. et al
    . Clinical outcomes of bacteremic pneumococcal pneumonia in the era of antibiotic resistance. Clin Infect Dis 2001;33:797-805.
    Abstract/FREE Full Text
  32. ↵
    1. Austrian R,
    2. Gold J
    . Pneumococcal bacteremia with special reference to bacteremic pneumococcal pneumonia. Ann Intern Med 1964;60:759-76.
    Abstract/FREE Full Text
  33. ↵
    1. Craig WA
    . Pharmacokinetic/pharmacodynamic parameters: rationale for antibacterial dosing of mice and men. Clin Infect Dis 1998;26:1-12.
    FREE Full Text
    1. Andes D
    . Pharmacokinetic and pharmacodynamic properties of antimicrobials in the therapy of respiratory tract infections. Curr Opin Infect Dis 2001;14:165-72.
    CrossRefMedlineWeb of Science
  35. ↵
    1. Craig WA
    . Interrelationship between pharmacokinetics and pharmacodynamics in determining dosage regimens for broad-spectrum cephalosporins. Diagn Microbiol Infect Dis 1995;22:89-96.
    CrossRefMedlineWeb of Science
  36. ↵
    1. Sacho H,
    2. Klugman KP,
    3. Koornhof HJ,
    4. Ruff P
    . Community-acquired pneumonia in an adult due to a multiply-resistant pneumococcus. J Infect 1987;14:188-9.
    CrossRefMedlineWeb of Science
    1. Buckingham SC,
    2. Brown SP,
    3. San Joaquin VH
    . Breakthrough bacteremia and meningitis during treatment with cephalosporins parenterally for pneumococcal pneumonia. J Pediatr 1998;132:174-6.
    CrossRefMedlineWeb of Science
  38. ↵
    1. Dowell SF,
    2. Smith T,
    3. Leversedge K,
    4. Snitzer J
    . Failure of treatment of pneumonia associated with highly resistant pneumococci in a child. Clin Infect Dis 1999;29:462-3.
    MedlineWeb of Science
  39. ↵
    1. Baddour LM,
    2. Smith EA
    . Multidrug resistance among Streptococcus pneumoniae in the United States. Clin Infect Dis 1999;29:224-5.
    FREE Full Text
  40. ↵
    1. Bartlett JG,
    2. Dowell SF,
    3. Mandell LA,
    4. File TM,
    5. Musher DM,
    6. Fine MJ
    . Practice guidelines for management of community-acquired pneumonia. Clin Infect Dis 2000;31:347-82.
    FREE Full Text
    1. Mandell LA,
    2. Marrie TJ,
    3. Grossman RF,
    4. Chow AW,
    5. Hyland RH
    . Canadian guidelines for the initial management of community-acquired pneumonia: an evidence-based update by the Canadian Infectious Diseases Society and the Canadian Thoracic Society. Canadian Community-Acquired Pneumonia Working Group. Clin Infect Dis 2000;31:383-421.
    FREE Full Text
  42. ↵
    1. Niederman MS,
    2. Mandell LA,
    3. Anzueto A,
    4. et al
    . Guidelines for the management of adults with community-acquired pneumonia: diagnosis, assessment of severity, antimicrobial therapy, and prevention. American Thoracic Society. Am J Respir Crit Care Med 2001;163:1730-54.
    FREE Full Text
  43. ↵
    1. Ho PL,
    2. Tse WS,
    3. Tsang KW,
    4. et al
    . Risk factors for acquisition of levofloxacin-resistant Streptococcus pneumoniae a case-control study. Clin Infect Dis 2001;32:701-7.
    Abstract/FREE Full Text
    1. Davidson R,
    2. Cavalcanti R,
    3. Brunton JL,
    4. et al
    . Resistance to levofloxacin and failure of treatment of pneumococcal pneumonia. N Engl J Med 2002;346:747-60.
    CrossRefMedlineWeb of Science
| Table of Contents

This Article

  1. Clin Infect Dis. (2003) 37 (2): 230-237. doi: 10.1086/377534
  1. AbstractFree
  2. » Full Text (HTML)Free
  3. Full Text (PDF)Free

Classifications

Share

  1. Email this article

Navigate This Article

Current Issue

  1. March 1, 2012 54 (5)
  1. Clinical Infectious Diseases
  1. Alert me to new issues

Most