&get_box_var;
ORIGINAL ARTICLE
The Effect of Macrolide Resistance on the Presentation and Outcome of Patients Hospitalized for Streptococcus pneumoniae Pneumonia Catia Cilloniz1, Richard K. Albert2,3, Adamanthia Liapikou4, Albert Gabarrus1, Ernesto Rangel5, Salvador Bello6, Francesc Marco7, Josep Mensa8, and Antoni Torres1 1
Department of Pneumology, Institut Clinic del Torax, ´ Hospital Clinic of Barcelona–Institut d’Investigacions Biomediques ` August Pi i Sunyer, University of Barcelona, Ciber de Enfermedades Respiratorias, Barcelona, Spain; 2Department of Medicine, Denver Health, Denver, Colorado; 3University of Colorado, Aurora, Colorado; 4Respiratory Department, Sotiria Chest Diseases Hospital, Athens, Greece; 5Facultad de Medicina, Universidad Autonoma de Nayarit, Tepic Nayarit, Mexico; 6Servicio de Neumologia, Hospital Universitario Miguel Servet, Zaragoza, Spain; and 7Department of Microbiology and 8Department of Infectious Disease, Hospital Clinic of Barcelona, Barcelona, Spain
Abstract Rationale: There are conflicting reports describing the effect of
macrolide resistance on the presentation and outcomes of patients with Streptococcus pneumoniae pneumonia. Objectives: We aimed to determine the effect of macrolide resistance on the presentation and outcomes of patients with pneumococcal pneumonia. Methods: We conducted a retrospective, observational study in
Patients with macrolide-resistant organisms were less likely to have bacteremia, pulmonary complications, and shock, and were less likely to require noninvasive mechanical ventilation. We found no increase in the incidence of acute renal failure, the frequency of intensive care unit admission, the need for invasive ventilatory support, the length of hospital stay, or the 30-day mortality in patients with (invasive or noninvasive) macrolide-resistant S. pneumoniae pneumonia, and no effect on outcomes as a function of whether treatment regimens did or did not comply with current guidelines.
the Hospital Clinic of Barcelona of all adult patients hospitalized with pneumonia who had positive cultures for S. pneumoniae from January 1, 2000 to December 31, 2013. Outcomes examined included bacteremia, pulmonary complications, acute renal failure, shock, intensive care unit admission, need for mechanical ventilation, length of hospital stay, and 30-day mortality.
Conclusions: We found no evidence suggesting that patients
Measurements and Main Results: Of 643 patients hospitalized
pneumoniae resistant to macrolide; pneumonia; pneumococcal pneumonia
for S. pneumoniae pneumonia, 139 (22%) were macrolide resistant.
Streptococcus pneumoniae is identified in 20–60% of patients requiring hospitalization for community-acquired pneumonia (CAP), making it the leading cause of CAP (1–3), and in 5.5–49% of patients hospitalized for healthcare-
hospitalized for macrolide-resistant S. pneumoniae pneumonia were more severely ill on presentation or had worse clinical outcomes if they were treated with guideline-compliant versus noncompliant regimens.
Keywords: community-acquired pneumonia; Streptococcus
associated pneumonia (HCAP) (1, 4, 5). The mortality of patients with S. pneumoniae pneumonia ranges from 6.4% in outpatients and in hospitalized patients who do not require treatment in an intensive care unit (ICU) to more than
35% in those requiring treatment in the ICU (6). Current Infectious Disease Society of America/American Thoracic Society (IDSA/ATS) guidelines recommend a macrolide antibiotic (azithromycin,
( Received in original form February 2, 2015; accepted in final form March 25, 2015 ) Supported by Ciber de Enfermedades Respiratorias (CibeRes CB06/06/0028), 2009 Support to Research Groups of Catalonia 911, and Institut d’Investigacions Biomediques ` August Pi i Sunyer. Author Contributions: A.T. had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design, A.T., R.K.A., and C.C. Acquisition of data, C.C., A.L., and E.R. Analysis and interpretation of data, C.C., R.K.A., F.M., S.B., J.M., and A.T. Drafting of the manuscript, A.T., R.K.A., and C.C. Critical revision of the manuscript for important intellectual content, A.T., C.C., and R.K.A. Statistical analysis, A.G. Study supervision, A.T., R.K.A., and C.C. Correspondence and requests for reprints should be addressed to Antoni Torres, M.D., Ph.D., Department of Pneumology, Hospital Clinic of Barcelona, Barcelona, Spain. E-mail: [email protected] This article has an online supplement, which is accessible from this issue’s table of contents at www.atsjournals.org Am J Respir Crit Care Med Vol 191, Iss 11, pp 1265–1272, Jun 1, 2015 Copyright © 2015 by the American Thoracic Society Originally Published in Press as DOI: 10.1164/rccm.201502-0212OC on March 25, 2015 Internet address: www.atsjournals.org
Cilloniz, Albert, Liapikou, et al.: Macrolide-Resistance Streptococcus pneumoniae Pneumonia
1265
ORIGINAL ARTICLE Study Design and Patients
At a Glance Commentary Scientific Knowledge on the Subject: There are conflicting reports
describing the effect of macrolide resistance on the presentation and outcomes of patients with Streptococcus pneumoniae pneumonia. What This Study Adds to the Field: Although the prevalence
of in vitro resistance to macrolide antibiotics is increasing, we found no evidence this resistance worsened outcomes in patients hospitalized for S. pneumoniae pneumonia. clarithromycin, or erythromycin) as first-line therapy for previously healthy patients who have no risk factors for drug-resistant S. pneumoniae infection (strong recommendation; level I evidence) and a combination of a macrolide and a b-lactam for patients who require hospitalization but not in the ICU. In patients with CAP who require ICU admission the guideline recommends using a combination of a b-lactam plus macrolide or a fluoroquinolone (5). The prevalence of macrolide resistance in S. pneumoniae is increasing with recent rates ranging from 18 to 35% (7–11). Although some studies link macrolide resistance with treatment failure in community-acquired respiratory tract infections (12–14), the effect of having macrolide-resistant S. pneumoniae on clinical outcomes of patients with pneumonia has not been clearly established. We performed a retrospective observational study of prospectively collected data to examine the effect of macrolide resistance on the outcomes of patients who were hospitalized for pneumonia caused by S. pneumoniae.
This was a retrospective observational study of data that were prospectively collected in the Hospital Clinic of Barcelona. Subjects included all adults admitted with S. pneumoniae pneumonia from January 1, 2000 to December 31, 2013, including those coming from nursing homes. Pneumonia was defined as the presence of a new infiltrate on a chest radiograph together with clinical symptoms that were suggestive of lower respiratory tract infection (e.g., fever, cough, sputum production, pleuritic chest pain). We excluded patients who were receiving an immunosuppressant, those taking more than 10 mg/day of prednisone, or cytotoxic therapy, and all patients known to have human immunodeficiency virus infection. Data Collection and Evaluation
At the time of hospital admission we recorded the patients’ age, sex, smoking history, alcohol use, illicit drug consumption, comorbidities, antibiotic treatment in the previous 30 days before hospital admission, whether they were receiving inhaled or systemic corticosteroids, clinical symptoms and signs, arterial blood gases, chest radiograph findings, a variety of laboratory tests, the results of diagnostic procedures (see later), the pneumonia severity index and CURB-65 score (consciousness, urea, respiratory rate, blood pressure, 65 year old) (15, 16), and the initial antibiotic therapy. Over the course of their hospitalization we recorded the length of hospital stay and the 30-day in-hospital mortality, and whether the patients required noninvasive and/or invasive ventilatory support, had any pulmonary complications, defined as multilobar infiltration, pleural effusions, or meeting criteria for the acute respiratory distress syndrome (17), or developed septic shock or acute renal failure.
Methods
Microbiologic Evaluation and Diagnostic Criteria
Ethics Statement
All patients had microbiologic examination of expectorated sputum, urine, two samples of blood, and nasopharyngeal swabs. Those who underwent a thoracentesis, intubation, or bronchoscopy also had microbiologic
The study was approved by the Ethics Committee of the Hospital Clinic of Barcelona (Barcelona, Spain; Register: 2009/ 5451). Written informed consent was waived because of the noninterventional design. 1266
examination of the pleural fluid, tracheobronchial aspirates, and/or bronchoalveolar lavage. Sputum and blood samples were obtained in the emergency department for bacterial culture before starting antibiotic therapy. Nasopharyngeal swabs were processed for respiratory virus detection. Urine was sent for S. pneumoniae and Legionella pneumophila antigen assessment within 24 hours after hospital admission. Sputum testing included Gram and Ziehl-Neelsen staining and culturing for bacterial, fungal, and mycobacterial pathogens. Blood samples for serology of atypical pathogens and respiratory viruses were collected at admission and between the third and sixth week thereafter. For the purpose of this study patients were considered to have S. pneumoniae pneumonia if S. pneumoniae was cultured from the blood, pleural fluid, tracheobronchial aspirates (at >105 CFU/ ml) or bronchoalveolar lavage (at >104 CFU/ml), or from sputum using standard microbiologic methods. Strains were initially screened for antimicrobial susceptibility by Sensititre (Trek Diagnostic Systems Ltd., West Sussex, UK). Penicillin and other antibiotic susceptibility were defined according to the 2012 break points by the Clinical Laboratory Standards Institute. For S. pneumoniae isolates, minimum inhibitory concentrations (MICs) were determined using the Sensititre for penicillin, cefotaxime, ceftriaxone, cefepime, imipenem, meropenem, erythromycin, clindamycin, levofloxacin, and vancomycin. Results were interpreted according to the 2012 Clinical and Laboratory Standards Institute criteria (performance standards for antimicrobial susceptibility testing, 22nd informational supplement, M100-S22; Clinical and Laboratory Standards Institute, Wayne, PA). All isolates were tested for antimicrobial susceptibilities using Clinical and Laboratory Standards Institute microdilution methods. Macrolide-susceptible S. pneumoniae was defined as an isolate with an erythromycin MIC less than or equal to 0.5 mg/L, intermediate resistance was defined as an erythromycin MIC of 1 mg/L, and resistance erythromycin was defined as an erythromycin MIC greater than or equal to 2 mg/L. S. pneumoniae was considered to be macrolide resistant when MICs were
American Journal of Respiratory and Critical Care Medicine Volume 191 Number 11 | June 1 2015
ORIGINAL ARTICLE Outcomes
5878 screened patients with CAP
Excluded (n=5235): Microbiological culture negative for antibiogram
643 patients analyzed
Macrolide-sensitive S. pneumoniae 504 patients (78.4%)
331 (65.6%) invasive cases (empyema + bacteremia)
Macrolide-resistant S. pneumoniae 139 (21.6%)
67 (48.2%) invasive cases (empyema + bacteremia)
Figure 1. Flow diagram of the selected population. CAP = community-acquired pneumonia.
greater than or equal to 1 mg/L (i.e., intermediate resistance or resistant). Data on macrolide-resistant specimens include those that showed resistance and those showing intermediate resistance. Appropriateness of empiric antibiotic treatment was defined according to IDSA/ ATS guidelines for management of CAP and HCAP (5, 18).
adjusted for multiple comparisons. A P value less than 0.05 was considered significant. All analyses were performed with IBM SPSS Statistics for Windows, Version 20.0 (IBM Corp., Armonk, NY).
Statistical Analysis
During the study period 5,878 patients were hospitalized with a diagnosis of CAP. Of these, 643 had one or more microbiologic studies that were positive for S. pneumoniae and 139 (22%) of these were macrolide resistant (Figure 1, Table 1). Their demographics and clinical characteristics are presented in Table 2. Because only 12 (1.8%) patients had intermediate resistance to macrolides their results were pooled with the 127 who had high level of resistance.
We report the mean and SD for continuous variables with normal distribution and the median (first quartile–third quartile) for those with nonnormal distribution and compared them using the t test or the nonparametric Mann-Whitney test, respectively. Categorical variables are presented as number of patients (percentage) and were compared using the chi-square test or Fisher exact test. All reported P values are two sided and not
Results Demographic and Clinical Variables on Presentation
Table 1. Results of Microbiologic Testing
Specimen Blood, n (%) Sputum, n (%) TBAS or BAL, n (%) Pleural fluid, n (%)
Isolates (n = 643) 356 264 51 34
(55) (41) (8) (5)
Macrolide Sensitive (n = 504; 78%) 294 197 38 26
(58) (39) (8) (5)
Macrolide Resistant (n = 139; 22%) 62 67 13 8
(45) (48) (9) (6)
Definition of abbreviations: BAL = bronchoalveolar lavage; TBAS = tracheobronchial aspirate.
By univariate analysis patients with macrolide-resistant S. pneumoniae pneumonia were more likely to have received antibiotics within the previous 30 days and more likely to have chronic obstructive pulmonary disease. Patients with macrolideresistant disease were less likely to have fever, bacteremia, pulmonary complications, or shock (Table 3) and were also less likely to require noninvasive ventilation (although the number of patients receiving noninvasive ventilation was too small for a meaningful comparison). We found no suggestion that patients with macrolide-resistant S. pneumoniae pneumonia presented with more severe disease or had worse clinical outcomes regardless of whether we did or did not exclude patients who died within the first 3 days of admission (Table 4) or if we restricted the analysis exclusively to patients with invasive disease (i.e., bacteremia and/or empyema) (see Table E1 in the online supplement). Outcomes Related to Antibiotic Treatment
Surprisingly, 129 patients (20%) were treated with a single antibiotic, a regimen that would be inconsistent with IDSA/ATS guidelines for treating hospitalized patients with either CAP or HCAP (Table 5). Despite this, we found no suggestion that patients receiving treatment that was inconsistent with these guidelines had worse clinical outcomes (Table 6). Outcomes of Patients with MacrolideResistant S. pneumoniae Pneumonia Who Received Combination Therapy That Did or Did Not Include a Macrolide
Of the 104 patients with macrolide-resistant S. pneumoniae pneumonia, 71 (68%) received a dual antibiotic regimen that included a macrolide and 33 (32%) did not. Despite their regimen containing only one antibiotic to which their organism was resistant, we found no difference in the outcomes of patients with macrolide-resistant versus macrolide-sensitive S. pneumoniae pneumonia (Table 7), with the exception that the frequency of ICU admission was higher in patients receiving dual therapy that did not include a macrolide.
Discussion The important findings of this study are that we could find no evidence of more severe
Cilloniz, Albert, Liapikou, et al.: Macrolide-Resistance Streptococcus pneumoniae Pneumonia
1267
ORIGINAL ARTICLE Table 2. Demographics and Clinical Characteristics on Admission Macrolide Sensitive (n = 504) Age, yr, mean 6 SD Male, sex, n (%) Current smoker, n (%) Current alcohol user, n (%) Nursing home resident, n (%) Previous antibiotics, n (%) Influenza vaccine, n (%) Pneumococcal vaccine, n (%) Previous inhaled steroids, n (%) Previous systemic steroids, n (%) Comorbidities, n (%) COPD Other chronic respiratory disease Chronic cardiovascular disease Diabetes mellitus Chronic neurologic disease Chronic renal disease Chronic liver disease Clinical symptoms, n (%) Fever, >388 C Cough Sputum production Pleuritic chest pain Dyspnea Confusion Vital signs, median (IQR) Heart rate, beats/min Respiratory rate, breaths/min Systolic blood pressure, mm Hg Diastolic blood pressure, mm Hg Laboratory tests, median (IQR) Creatinine, mg/ml C-reactive protein, mg/dl White blood cell count, 109/L PaO2/FIO2, mm Hg PSI risk score, n (%) I II III IV V CURB-65 risk class, n (%) 0 1 2 3 4 5
Macrolide Resistant (n = 139)
P Value
63 6 18 330 (66) 179 (36) 107 (21) 15 (3) 51 (11) 107 (34) 52 (10) 95 (19) 19 (5)
67 6 18 87 (63) 38 (28) 25 (19) 7 (5) 26 (19) 32 (40) 10 (7) 35 (26) 6 (6)
0.005 0.51 0.071 0.46 0.25 0.008 0.29 0.27 0.096 0.72
110 121 61 90 53 29 39
(22) (24) (12) (18) (11) (6) (8)
44 41 23 24 23 7 12
(32) (29) (17) (17) (17) (5) (9)
0.015 0.18 0.18 0.81 0.063 0.72 0.75
429 418 341 269 359 98
(86) (84) (71) (55) (73) (20)
108 123 108 75 109 26
(78) (89) (79) (55) (78) (19)
0.013 0.17 0.057 .0.99 0.17 0.82
104 28 128 70
(90–120) (23–32) (110–145) (62–80)
100 28 128 70
(88–113) (24–32) (112–145) (60–80)
0.069 0.64 0.47 0.79
1.1 24 15.4 271
(0.9–1.5) (13–32) (10.1–20.4) (229–310)
1.1 22 13.8 252
(0.9–1.4) (11–30) (9.9–18.6) (219–305)
0.29 0.16 0.13 0.057 0.10
4 131 105 174 90
(1) (26) (21) (35) (18)
0 25 26 63 25
(0) (18) (19) (45) (18)
90 164 131 74 27 2
(18) (34) (27) (15) (6) (0.4)
22 37 47 21 6 3
(16) (27) (35) (15) (4) (2)
0.13
Definition of abbreviations: COPD = chronic obstructive pulmonary disease; CURB-65 = consciousness, urea, respiratory rate, blood pressure, 65 year old; IQR = interquartile range; PSI = pneumonia severity index. Percentages are calculated on nonmissing data. Bold values indicate statistical significance.
presentations or worse clinical outcomes in patients who were admitted to the hospital with roentgenographically proven pneumonia caused by S. pneumoniae pneumonia as a function of whether the organisms cultured were sensitive or resistant to macrolide antibiotics or if the patients had invasive or noninvasive disease. We also found no evidence of worse clinical outcomes in patients who 1268
were treated with regimens that were consistent versus inconsistent with current guidelines for treating CAP or HCAP. Literature on the effect of macrolide resistance on outcomes of patients with S. pneumoniae infections is conflicting. Twenty years ago Moreno and colleagues (19) found no difference in hospital mortality in patients with macrolidesensitive versus -resistant S. pneumoniae
(14% vs. 18% in those with macrolidesensitive and -resistant infections, respectively; P = 0.8). Two subsequent studies found trends toward an increased mortality in patients with macrolideresistant pneumococcal disease that were not statistically significant (20, 21) but Song and colleagues (22) did not (P = 0.6 or 0.9 for patients with pneumonia severity index 1–5 and 4 or 5, respectively).
American Journal of Respiratory and Critical Care Medicine Volume 191 Number 11 | June 1 2015
ORIGINAL ARTICLE Table 3. Outcomes According to Macrolide Sensitivity Macrolide Sensitive (n = 504) Bacteremia, n (%) Days of hospital stay, median (IQR) 30-d in-hospital mortality, n (%) ICU admission, n (%) Mechanical ventilation,* n (%) None Noninvasive Invasive Pulmonary complications,† n (%) Multilobar infiltration Pleural effusion ARDS Acute renal failure, n (%) Shock, n (%)
299 8 40 145
(60) (5–12) (8) (29)
387 22 50 218 149 98 28 153 67
(84) (5) (11) (44) (30) (20) (6) (31) (14)
Macrolide Resistant (n = 139) 65 8 13 35 106 1 19 45 32 20 6 41 10
P Value
(47) (4–14) (9) (25)
0.009 0.96 0.59 0.40 0.064 0.96 0.041 0.20 0.017 0.13 0.16 0.55 0.84 0.050
(84) (1) (15) (32) (23) (14) (5) (30) (7)
Definition of abbreviations: ARDS = acute respiratory distress syndrome; ICU = intensive care unit; IQR = interquartile range. Percentages are calculated on nonmissing data. Bold values indicate statistical significance. *Patients who received initially noninvasive ventilation but needed subsequent intubation were included in the invasive mechanical ventilation group. † Patients could have more than one pulmonary complication.
In an observational study Asadi and colleagues (23) found that outpatients with CAP had a lower mortality if they were treated with regimens that were consistent with published guidelines compared with those whose regimens were not (6% vs. 1%, respectively; odds ratio, 0.23; 95% confidence interval, 0.09–0.59; P = 0.002). Within the group receiving treatment that was concordant with guidelines, those receiving macrolides were less likely to die within 30 days (64% vs. 0.2%; odds ratio, 2.3; 95% confidence interval, 0.09–0.86; P = 0.03). Other observational studies also find that mortality is lower in inpatients with CAP and pneumococcal bacteremia who are treated with macrolides
(21, 24, 25) but no benefit was seen if data are restricted to randomized controlled trials (24). Baddour and colleagues (26) found no difference in the mortality of patients treated with combination antibiotic therapy versus monotherapy unless the patients were critically ill. We found that patients hospitalized with S. pneumoniae pneumonia who were treated with guideline-compliant regimens had lengths of hospital stay than those treated with regimens that were noncompliant but the patients treated with guideline-consistent regimens were more likely to have bacteremia on admission, multilobar infiltration, acute respiratory distress syndrome, and acute renal failure (Table 6).
Several observational studies and a recent randomized controlled trial have found improved outcomes in patients with CAP if their treatment regimens included a macrolide antibiotic (27–29). We found no difference in outcomes in patients whose regimens did or did not include a macrolide (Table 7), except that patients with S. pneumoniae resistant to macrolide had less need for ICU admission. This finding opens again the question of the potential antiinflammatory effect of macrolide (30, 31). Our study has a number of limitations. First, because the data were collected from a single academic teaching hospital in Spain the results might not generalize to other
Table 4. Outcomes According to Macrolide Sensitivity Excluding Patients Who Died within the First 3 Days of Admission Macrolide Sensitive (n = 493) Bacteremia, n (%) Days of hospital stay, median (IQR) 30-d in-hospital mortality, n (%) ICU admission, n (%) Mechanical ventilation,* n (%) None Noninvasive Invasive Pulmonary complications,† n (%) Multilobar infiltration Pleural effusion ARDS Acute renal failure, n (%) Shock, n (%)
295 8 32 140
(60) (5–12) (6) (28)
385 20 47 211 143 96 24 147 63
(85) (4) (10) (43) (29) (20) (5) (30) (13)
Macrolide Resistant (n = 133) 62 8 9 32 105 1 16 42 29 18 4 37 7
(47) (5–14) (7) (24) (86) (1) (13) (31) (21) (13) (3) (28) (5)
P Value 0.006 .0.99 0.93 0.29 0.13 0.81 0.060 0.39 0.013 0.089 0.097 0.34 0.61 0.014
Definition of abbreviations: ARDS = acute respiratory distress syndrome; ICU = intensive care unit; IQR = interquartile range. Percentages are calculated on nonmissing data. Bold values indicate statistical significance. *Patients who received initially noninvasive ventilation but needed subsequent intubation were included in the invasive mechanical ventilation group. † Patients could have more than one pulmonary complication.
Cilloniz, Albert, Liapikou, et al.: Macrolide-Resistance Streptococcus pneumoniae Pneumonia
1269
ORIGINAL ARTICLE for multiple comparisons. Fifth, because we only encountered three patients with macrolide-resistant disease who were treated with macrolide monotherapy we cannot comment on the use of this regimen in this setting, although macrolide monotherapy would not be an appropriate choice for treating hospitalized patients.
Table 5. Antibiotic Regimens Macrolide Sensitive (n = 504) Monotherapy, n (%) Quinolone b-Lactam Macrolide Other Combination, n (%) b-Lactam 1 macrolide b-Lactam 1 quinolone Macrolide 1 quinolone Other Guideline-consistent regimens, n (%) Guideline-inconsistent regimens, n (%)
95 53 35 4 3 405 229 114 8 54 346 154
Macrolide Resistant (n = 139)
(19) (11) (7) (1) (1) (81) (46) (23) (2) (11) (69) (31)
34 22 7 3 2 104 69 22 1 12 91 47
(25) (16) (5) (2) (1) (75) (50) (16) (1) (9) (66) (34)
Conclusions
We found no evidence suggesting that patients hospitalized with culture- and roentgenographically proven, invasive or noninvasive S. pneumoniae pneumonia were more severely ill at presentation or had worse outcomes if their organism was resistant versus sensitive to macrolide antibiotics and/or if they were treated with guideline-compliant versus noncompliant regimens. Although the prevalence of in vitro resistance to macrolide antibiotics is increasing, we found no evidence that this resistance worsened outcomes in patients hospitalized for S. pneumoniae pneumonia. n
Percentages are calculated on nonmissing data.
patients admitted to other types of hospitals in other countries. Second, although we found no difference in the outcomes in patients with macrolideresistant versus -sensitive organisms, we could only analyze 139 patients with macrolide-resistant organisms and this sample size may result in a large type II error. Our sample size, however, is larger than many previous studies of macrolideresistant S. pneumoniae reported in the literature (14, 19, 20, 22, 32–34). Third, we did not genotype the S. pneumoniae
isolates to determine their specific mechanism of macrolide resistance and our failure to find any differences in outcomes could have resulted from a high fraction of the macrolide-resistant organisms having the ermB gene. A Spanish study found that 89.9% of macrolide-resistant pneumococci carried the ermB gene (35) but Daneman and colleagues (36) found equal representation of the mefA and ermB genes in their series of macrolide failures. Fourth, there were no adjustments made
Author disclosures are available with the text of this article at www.atsjournals.org. Acknowledgment: The authors are indebted to all medical and nursing colleagues for their assistance and cooperation in this study.
Table 6. Outcomes according to Appropriateness of Treatment and Macrolide Sensitivity Guideline-Consistent Regimens (n = 437)
Bacteremia, n (%) Days of hospital stay, median (IQR) 30-d in-hospital mortality, n (%) ICU admission, n (%) Mechanical ventilation,x n (%) None Noninvasive Invasive Pulmonary complications,jj n (%) Multilobar infiltration Pleural effusion ARDS Acute renal failure, n (%) Shock, n (%)
Macrolide Sensitive (n = 346)
Macrolide Resistant (n = 91)
Total (n = 437)
217 (63) 8 (5–12)
46 (51) 9 (5–14)
263 (61) 8 (5–13)
P Value
Guideline-Inconsistent Regimens (n = 201) Macrolide Sensitive (n = 154)
0.039 0.81
81 (53) 6.5 (4–11)
Macrolide Resistant (n = 47)
Total (n = 201)
19 (41) 6 (4–13)
100 (51) 6 (4–12)
P Value†
P Value‡
P Value
P Value*
0.15 0.74
0.040 0.005
0.28 0.10
0.017 0.001
27 (8)
6 (7)
33 (8)
0.70
13 (8)
6 (13)
19 (9)
0.38
0.81
0.22
0.42
109 (32)
23 (25)
132 (30)
0.25 0.046
36 (24)
11 (23)
47 (24)
0.99 0.58
0.071
0.81
0.081 0.040
265 21 32 166
(83) (7) (10) (49)
71 0 11 28
(87) (0) (13) (31)
336 21 43 194
(84) (5) (11) (45)
0.48 0.017 0.38 0.002
119 1 18 50
(86) (1) (13) (32)
35 1 7 16
(81) (2) (16) (34)
154 2 25 66
(85) (1) (14) (33)
0.44 0.38 0.59 0.84
0.44 0.007 0.35 0.001
0.44 0.17 0.66 0.70
0.74 0.018 0.29 0.004
116 69 24 119 47
(34) (20) (7) (35) (14)
19 13 5 33 6
(21) (14) (6) (37) (7)
135 82 29 152 53
(31) (19) (7) (36) (12)
0.020 0.20 0.58 0.80 0.061
32 28 4 34 20
(21) (18) (3) (22) (13)
12 7 1 7 3
(26) (15) (2) (15) (7)
44 35 5 41 23
(22) (18) (3) (21) (12)
0.49 0.59 0.88 0.31 0.24
0.040 0.63 0.052 0.004 0.79
0.54 0.92 0.40 0.009 0.99
0.019 0.67 0.033
ORIGINAL ARTICLE
The Effect of Macrolide Resistance on the Presentation and Outcome of Patients Hospitalized for Streptococcus pneumoniae Pneumonia Catia Cilloniz1, Richard K. Albert2,3, Adamanthia Liapikou4, Albert Gabarrus1, Ernesto Rangel5, Salvador Bello6, Francesc Marco7, Josep Mensa8, and Antoni Torres1 1
Department of Pneumology, Institut Clinic del Torax, ´ Hospital Clinic of Barcelona–Institut d’Investigacions Biomediques ` August Pi i Sunyer, University of Barcelona, Ciber de Enfermedades Respiratorias, Barcelona, Spain; 2Department of Medicine, Denver Health, Denver, Colorado; 3University of Colorado, Aurora, Colorado; 4Respiratory Department, Sotiria Chest Diseases Hospital, Athens, Greece; 5Facultad de Medicina, Universidad Autonoma de Nayarit, Tepic Nayarit, Mexico; 6Servicio de Neumologia, Hospital Universitario Miguel Servet, Zaragoza, Spain; and 7Department of Microbiology and 8Department of Infectious Disease, Hospital Clinic of Barcelona, Barcelona, Spain
Abstract Rationale: There are conflicting reports describing the effect of
macrolide resistance on the presentation and outcomes of patients with Streptococcus pneumoniae pneumonia. Objectives: We aimed to determine the effect of macrolide resistance on the presentation and outcomes of patients with pneumococcal pneumonia. Methods: We conducted a retrospective, observational study in
Patients with macrolide-resistant organisms were less likely to have bacteremia, pulmonary complications, and shock, and were less likely to require noninvasive mechanical ventilation. We found no increase in the incidence of acute renal failure, the frequency of intensive care unit admission, the need for invasive ventilatory support, the length of hospital stay, or the 30-day mortality in patients with (invasive or noninvasive) macrolide-resistant S. pneumoniae pneumonia, and no effect on outcomes as a function of whether treatment regimens did or did not comply with current guidelines.
the Hospital Clinic of Barcelona of all adult patients hospitalized with pneumonia who had positive cultures for S. pneumoniae from January 1, 2000 to December 31, 2013. Outcomes examined included bacteremia, pulmonary complications, acute renal failure, shock, intensive care unit admission, need for mechanical ventilation, length of hospital stay, and 30-day mortality.
Conclusions: We found no evidence suggesting that patients
Measurements and Main Results: Of 643 patients hospitalized
pneumoniae resistant to macrolide; pneumonia; pneumococcal pneumonia
for S. pneumoniae pneumonia, 139 (22%) were macrolide resistant.
Streptococcus pneumoniae is identified in 20–60% of patients requiring hospitalization for community-acquired pneumonia (CAP), making it the leading cause of CAP (1–3), and in 5.5–49% of patients hospitalized for healthcare-
hospitalized for macrolide-resistant S. pneumoniae pneumonia were more severely ill on presentation or had worse clinical outcomes if they were treated with guideline-compliant versus noncompliant regimens.
Keywords: community-acquired pneumonia; Streptococcus
associated pneumonia (HCAP) (1, 4, 5). The mortality of patients with S. pneumoniae pneumonia ranges from 6.4% in outpatients and in hospitalized patients who do not require treatment in an intensive care unit (ICU) to more than
35% in those requiring treatment in the ICU (6). Current Infectious Disease Society of America/American Thoracic Society (IDSA/ATS) guidelines recommend a macrolide antibiotic (azithromycin,
( Received in original form February 2, 2015; accepted in final form March 25, 2015 ) Supported by Ciber de Enfermedades Respiratorias (CibeRes CB06/06/0028), 2009 Support to Research Groups of Catalonia 911, and Institut d’Investigacions Biomediques ` August Pi i Sunyer. Author Contributions: A.T. had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design, A.T., R.K.A., and C.C. Acquisition of data, C.C., A.L., and E.R. Analysis and interpretation of data, C.C., R.K.A., F.M., S.B., J.M., and A.T. Drafting of the manuscript, A.T., R.K.A., and C.C. Critical revision of the manuscript for important intellectual content, A.T., C.C., and R.K.A. Statistical analysis, A.G. Study supervision, A.T., R.K.A., and C.C. Correspondence and requests for reprints should be addressed to Antoni Torres, M.D., Ph.D., Department of Pneumology, Hospital Clinic of Barcelona, Barcelona, Spain. E-mail: [email protected] This article has an online supplement, which is accessible from this issue’s table of contents at www.atsjournals.org Am J Respir Crit Care Med Vol 191, Iss 11, pp 1265–1272, Jun 1, 2015 Copyright © 2015 by the American Thoracic Society Originally Published in Press as DOI: 10.1164/rccm.201502-0212OC on March 25, 2015 Internet address: www.atsjournals.org
Cilloniz, Albert, Liapikou, et al.: Macrolide-Resistance Streptococcus pneumoniae Pneumonia
1265
ORIGINAL ARTICLE Study Design and Patients
At a Glance Commentary Scientific Knowledge on the Subject: There are conflicting reports
describing the effect of macrolide resistance on the presentation and outcomes of patients with Streptococcus pneumoniae pneumonia. What This Study Adds to the Field: Although the prevalence
of in vitro resistance to macrolide antibiotics is increasing, we found no evidence this resistance worsened outcomes in patients hospitalized for S. pneumoniae pneumonia. clarithromycin, or erythromycin) as first-line therapy for previously healthy patients who have no risk factors for drug-resistant S. pneumoniae infection (strong recommendation; level I evidence) and a combination of a macrolide and a b-lactam for patients who require hospitalization but not in the ICU. In patients with CAP who require ICU admission the guideline recommends using a combination of a b-lactam plus macrolide or a fluoroquinolone (5). The prevalence of macrolide resistance in S. pneumoniae is increasing with recent rates ranging from 18 to 35% (7–11). Although some studies link macrolide resistance with treatment failure in community-acquired respiratory tract infections (12–14), the effect of having macrolide-resistant S. pneumoniae on clinical outcomes of patients with pneumonia has not been clearly established. We performed a retrospective observational study of prospectively collected data to examine the effect of macrolide resistance on the outcomes of patients who were hospitalized for pneumonia caused by S. pneumoniae.
This was a retrospective observational study of data that were prospectively collected in the Hospital Clinic of Barcelona. Subjects included all adults admitted with S. pneumoniae pneumonia from January 1, 2000 to December 31, 2013, including those coming from nursing homes. Pneumonia was defined as the presence of a new infiltrate on a chest radiograph together with clinical symptoms that were suggestive of lower respiratory tract infection (e.g., fever, cough, sputum production, pleuritic chest pain). We excluded patients who were receiving an immunosuppressant, those taking more than 10 mg/day of prednisone, or cytotoxic therapy, and all patients known to have human immunodeficiency virus infection. Data Collection and Evaluation
At the time of hospital admission we recorded the patients’ age, sex, smoking history, alcohol use, illicit drug consumption, comorbidities, antibiotic treatment in the previous 30 days before hospital admission, whether they were receiving inhaled or systemic corticosteroids, clinical symptoms and signs, arterial blood gases, chest radiograph findings, a variety of laboratory tests, the results of diagnostic procedures (see later), the pneumonia severity index and CURB-65 score (consciousness, urea, respiratory rate, blood pressure, 65 year old) (15, 16), and the initial antibiotic therapy. Over the course of their hospitalization we recorded the length of hospital stay and the 30-day in-hospital mortality, and whether the patients required noninvasive and/or invasive ventilatory support, had any pulmonary complications, defined as multilobar infiltration, pleural effusions, or meeting criteria for the acute respiratory distress syndrome (17), or developed septic shock or acute renal failure.
Methods
Microbiologic Evaluation and Diagnostic Criteria
Ethics Statement
All patients had microbiologic examination of expectorated sputum, urine, two samples of blood, and nasopharyngeal swabs. Those who underwent a thoracentesis, intubation, or bronchoscopy also had microbiologic
The study was approved by the Ethics Committee of the Hospital Clinic of Barcelona (Barcelona, Spain; Register: 2009/ 5451). Written informed consent was waived because of the noninterventional design. 1266
examination of the pleural fluid, tracheobronchial aspirates, and/or bronchoalveolar lavage. Sputum and blood samples were obtained in the emergency department for bacterial culture before starting antibiotic therapy. Nasopharyngeal swabs were processed for respiratory virus detection. Urine was sent for S. pneumoniae and Legionella pneumophila antigen assessment within 24 hours after hospital admission. Sputum testing included Gram and Ziehl-Neelsen staining and culturing for bacterial, fungal, and mycobacterial pathogens. Blood samples for serology of atypical pathogens and respiratory viruses were collected at admission and between the third and sixth week thereafter. For the purpose of this study patients were considered to have S. pneumoniae pneumonia if S. pneumoniae was cultured from the blood, pleural fluid, tracheobronchial aspirates (at >105 CFU/ ml) or bronchoalveolar lavage (at >104 CFU/ml), or from sputum using standard microbiologic methods. Strains were initially screened for antimicrobial susceptibility by Sensititre (Trek Diagnostic Systems Ltd., West Sussex, UK). Penicillin and other antibiotic susceptibility were defined according to the 2012 break points by the Clinical Laboratory Standards Institute. For S. pneumoniae isolates, minimum inhibitory concentrations (MICs) were determined using the Sensititre for penicillin, cefotaxime, ceftriaxone, cefepime, imipenem, meropenem, erythromycin, clindamycin, levofloxacin, and vancomycin. Results were interpreted according to the 2012 Clinical and Laboratory Standards Institute criteria (performance standards for antimicrobial susceptibility testing, 22nd informational supplement, M100-S22; Clinical and Laboratory Standards Institute, Wayne, PA). All isolates were tested for antimicrobial susceptibilities using Clinical and Laboratory Standards Institute microdilution methods. Macrolide-susceptible S. pneumoniae was defined as an isolate with an erythromycin MIC less than or equal to 0.5 mg/L, intermediate resistance was defined as an erythromycin MIC of 1 mg/L, and resistance erythromycin was defined as an erythromycin MIC greater than or equal to 2 mg/L. S. pneumoniae was considered to be macrolide resistant when MICs were
American Journal of Respiratory and Critical Care Medicine Volume 191 Number 11 | June 1 2015
ORIGINAL ARTICLE Outcomes
5878 screened patients with CAP
Excluded (n=5235): Microbiological culture negative for antibiogram
643 patients analyzed
Macrolide-sensitive S. pneumoniae 504 patients (78.4%)
331 (65.6%) invasive cases (empyema + bacteremia)
Macrolide-resistant S. pneumoniae 139 (21.6%)
67 (48.2%) invasive cases (empyema + bacteremia)
Figure 1. Flow diagram of the selected population. CAP = community-acquired pneumonia.
greater than or equal to 1 mg/L (i.e., intermediate resistance or resistant). Data on macrolide-resistant specimens include those that showed resistance and those showing intermediate resistance. Appropriateness of empiric antibiotic treatment was defined according to IDSA/ ATS guidelines for management of CAP and HCAP (5, 18).
adjusted for multiple comparisons. A P value less than 0.05 was considered significant. All analyses were performed with IBM SPSS Statistics for Windows, Version 20.0 (IBM Corp., Armonk, NY).
Statistical Analysis
During the study period 5,878 patients were hospitalized with a diagnosis of CAP. Of these, 643 had one or more microbiologic studies that were positive for S. pneumoniae and 139 (22%) of these were macrolide resistant (Figure 1, Table 1). Their demographics and clinical characteristics are presented in Table 2. Because only 12 (1.8%) patients had intermediate resistance to macrolides their results were pooled with the 127 who had high level of resistance.
We report the mean and SD for continuous variables with normal distribution and the median (first quartile–third quartile) for those with nonnormal distribution and compared them using the t test or the nonparametric Mann-Whitney test, respectively. Categorical variables are presented as number of patients (percentage) and were compared using the chi-square test or Fisher exact test. All reported P values are two sided and not
Results Demographic and Clinical Variables on Presentation
Table 1. Results of Microbiologic Testing
Specimen Blood, n (%) Sputum, n (%) TBAS or BAL, n (%) Pleural fluid, n (%)
Isolates (n = 643) 356 264 51 34
(55) (41) (8) (5)
Macrolide Sensitive (n = 504; 78%) 294 197 38 26
(58) (39) (8) (5)
Macrolide Resistant (n = 139; 22%) 62 67 13 8
(45) (48) (9) (6)
Definition of abbreviations: BAL = bronchoalveolar lavage; TBAS = tracheobronchial aspirate.
By univariate analysis patients with macrolide-resistant S. pneumoniae pneumonia were more likely to have received antibiotics within the previous 30 days and more likely to have chronic obstructive pulmonary disease. Patients with macrolideresistant disease were less likely to have fever, bacteremia, pulmonary complications, or shock (Table 3) and were also less likely to require noninvasive ventilation (although the number of patients receiving noninvasive ventilation was too small for a meaningful comparison). We found no suggestion that patients with macrolide-resistant S. pneumoniae pneumonia presented with more severe disease or had worse clinical outcomes regardless of whether we did or did not exclude patients who died within the first 3 days of admission (Table 4) or if we restricted the analysis exclusively to patients with invasive disease (i.e., bacteremia and/or empyema) (see Table E1 in the online supplement). Outcomes Related to Antibiotic Treatment
Surprisingly, 129 patients (20%) were treated with a single antibiotic, a regimen that would be inconsistent with IDSA/ATS guidelines for treating hospitalized patients with either CAP or HCAP (Table 5). Despite this, we found no suggestion that patients receiving treatment that was inconsistent with these guidelines had worse clinical outcomes (Table 6). Outcomes of Patients with MacrolideResistant S. pneumoniae Pneumonia Who Received Combination Therapy That Did or Did Not Include a Macrolide
Of the 104 patients with macrolide-resistant S. pneumoniae pneumonia, 71 (68%) received a dual antibiotic regimen that included a macrolide and 33 (32%) did not. Despite their regimen containing only one antibiotic to which their organism was resistant, we found no difference in the outcomes of patients with macrolide-resistant versus macrolide-sensitive S. pneumoniae pneumonia (Table 7), with the exception that the frequency of ICU admission was higher in patients receiving dual therapy that did not include a macrolide.
Discussion The important findings of this study are that we could find no evidence of more severe
Cilloniz, Albert, Liapikou, et al.: Macrolide-Resistance Streptococcus pneumoniae Pneumonia
1267
ORIGINAL ARTICLE Table 2. Demographics and Clinical Characteristics on Admission Macrolide Sensitive (n = 504) Age, yr, mean 6 SD Male, sex, n (%) Current smoker, n (%) Current alcohol user, n (%) Nursing home resident, n (%) Previous antibiotics, n (%) Influenza vaccine, n (%) Pneumococcal vaccine, n (%) Previous inhaled steroids, n (%) Previous systemic steroids, n (%) Comorbidities, n (%) COPD Other chronic respiratory disease Chronic cardiovascular disease Diabetes mellitus Chronic neurologic disease Chronic renal disease Chronic liver disease Clinical symptoms, n (%) Fever, >388 C Cough Sputum production Pleuritic chest pain Dyspnea Confusion Vital signs, median (IQR) Heart rate, beats/min Respiratory rate, breaths/min Systolic blood pressure, mm Hg Diastolic blood pressure, mm Hg Laboratory tests, median (IQR) Creatinine, mg/ml C-reactive protein, mg/dl White blood cell count, 109/L PaO2/FIO2, mm Hg PSI risk score, n (%) I II III IV V CURB-65 risk class, n (%) 0 1 2 3 4 5
Macrolide Resistant (n = 139)
P Value
63 6 18 330 (66) 179 (36) 107 (21) 15 (3) 51 (11) 107 (34) 52 (10) 95 (19) 19 (5)
67 6 18 87 (63) 38 (28) 25 (19) 7 (5) 26 (19) 32 (40) 10 (7) 35 (26) 6 (6)
0.005 0.51 0.071 0.46 0.25 0.008 0.29 0.27 0.096 0.72
110 121 61 90 53 29 39
(22) (24) (12) (18) (11) (6) (8)
44 41 23 24 23 7 12
(32) (29) (17) (17) (17) (5) (9)
0.015 0.18 0.18 0.81 0.063 0.72 0.75
429 418 341 269 359 98
(86) (84) (71) (55) (73) (20)
108 123 108 75 109 26
(78) (89) (79) (55) (78) (19)
0.013 0.17 0.057 .0.99 0.17 0.82
104 28 128 70
(90–120) (23–32) (110–145) (62–80)
100 28 128 70
(88–113) (24–32) (112–145) (60–80)
0.069 0.64 0.47 0.79
1.1 24 15.4 271
(0.9–1.5) (13–32) (10.1–20.4) (229–310)
1.1 22 13.8 252
(0.9–1.4) (11–30) (9.9–18.6) (219–305)
0.29 0.16 0.13 0.057 0.10
4 131 105 174 90
(1) (26) (21) (35) (18)
0 25 26 63 25
(0) (18) (19) (45) (18)
90 164 131 74 27 2
(18) (34) (27) (15) (6) (0.4)
22 37 47 21 6 3
(16) (27) (35) (15) (4) (2)
0.13
Definition of abbreviations: COPD = chronic obstructive pulmonary disease; CURB-65 = consciousness, urea, respiratory rate, blood pressure, 65 year old; IQR = interquartile range; PSI = pneumonia severity index. Percentages are calculated on nonmissing data. Bold values indicate statistical significance.
presentations or worse clinical outcomes in patients who were admitted to the hospital with roentgenographically proven pneumonia caused by S. pneumoniae pneumonia as a function of whether the organisms cultured were sensitive or resistant to macrolide antibiotics or if the patients had invasive or noninvasive disease. We also found no evidence of worse clinical outcomes in patients who 1268
were treated with regimens that were consistent versus inconsistent with current guidelines for treating CAP or HCAP. Literature on the effect of macrolide resistance on outcomes of patients with S. pneumoniae infections is conflicting. Twenty years ago Moreno and colleagues (19) found no difference in hospital mortality in patients with macrolidesensitive versus -resistant S. pneumoniae
(14% vs. 18% in those with macrolidesensitive and -resistant infections, respectively; P = 0.8). Two subsequent studies found trends toward an increased mortality in patients with macrolideresistant pneumococcal disease that were not statistically significant (20, 21) but Song and colleagues (22) did not (P = 0.6 or 0.9 for patients with pneumonia severity index 1–5 and 4 or 5, respectively).
American Journal of Respiratory and Critical Care Medicine Volume 191 Number 11 | June 1 2015
ORIGINAL ARTICLE Table 3. Outcomes According to Macrolide Sensitivity Macrolide Sensitive (n = 504) Bacteremia, n (%) Days of hospital stay, median (IQR) 30-d in-hospital mortality, n (%) ICU admission, n (%) Mechanical ventilation,* n (%) None Noninvasive Invasive Pulmonary complications,† n (%) Multilobar infiltration Pleural effusion ARDS Acute renal failure, n (%) Shock, n (%)
299 8 40 145
(60) (5–12) (8) (29)
387 22 50 218 149 98 28 153 67
(84) (5) (11) (44) (30) (20) (6) (31) (14)
Macrolide Resistant (n = 139) 65 8 13 35 106 1 19 45 32 20 6 41 10
P Value
(47) (4–14) (9) (25)
0.009 0.96 0.59 0.40 0.064 0.96 0.041 0.20 0.017 0.13 0.16 0.55 0.84 0.050
(84) (1) (15) (32) (23) (14) (5) (30) (7)
Definition of abbreviations: ARDS = acute respiratory distress syndrome; ICU = intensive care unit; IQR = interquartile range. Percentages are calculated on nonmissing data. Bold values indicate statistical significance. *Patients who received initially noninvasive ventilation but needed subsequent intubation were included in the invasive mechanical ventilation group. † Patients could have more than one pulmonary complication.
In an observational study Asadi and colleagues (23) found that outpatients with CAP had a lower mortality if they were treated with regimens that were consistent with published guidelines compared with those whose regimens were not (6% vs. 1%, respectively; odds ratio, 0.23; 95% confidence interval, 0.09–0.59; P = 0.002). Within the group receiving treatment that was concordant with guidelines, those receiving macrolides were less likely to die within 30 days (64% vs. 0.2%; odds ratio, 2.3; 95% confidence interval, 0.09–0.86; P = 0.03). Other observational studies also find that mortality is lower in inpatients with CAP and pneumococcal bacteremia who are treated with macrolides
(21, 24, 25) but no benefit was seen if data are restricted to randomized controlled trials (24). Baddour and colleagues (26) found no difference in the mortality of patients treated with combination antibiotic therapy versus monotherapy unless the patients were critically ill. We found that patients hospitalized with S. pneumoniae pneumonia who were treated with guideline-compliant regimens had lengths of hospital stay than those treated with regimens that were noncompliant but the patients treated with guideline-consistent regimens were more likely to have bacteremia on admission, multilobar infiltration, acute respiratory distress syndrome, and acute renal failure (Table 6).
Several observational studies and a recent randomized controlled trial have found improved outcomes in patients with CAP if their treatment regimens included a macrolide antibiotic (27–29). We found no difference in outcomes in patients whose regimens did or did not include a macrolide (Table 7), except that patients with S. pneumoniae resistant to macrolide had less need for ICU admission. This finding opens again the question of the potential antiinflammatory effect of macrolide (30, 31). Our study has a number of limitations. First, because the data were collected from a single academic teaching hospital in Spain the results might not generalize to other
Table 4. Outcomes According to Macrolide Sensitivity Excluding Patients Who Died within the First 3 Days of Admission Macrolide Sensitive (n = 493) Bacteremia, n (%) Days of hospital stay, median (IQR) 30-d in-hospital mortality, n (%) ICU admission, n (%) Mechanical ventilation,* n (%) None Noninvasive Invasive Pulmonary complications,† n (%) Multilobar infiltration Pleural effusion ARDS Acute renal failure, n (%) Shock, n (%)
295 8 32 140
(60) (5–12) (6) (28)
385 20 47 211 143 96 24 147 63
(85) (4) (10) (43) (29) (20) (5) (30) (13)
Macrolide Resistant (n = 133) 62 8 9 32 105 1 16 42 29 18 4 37 7
(47) (5–14) (7) (24) (86) (1) (13) (31) (21) (13) (3) (28) (5)
P Value 0.006 .0.99 0.93 0.29 0.13 0.81 0.060 0.39 0.013 0.089 0.097 0.34 0.61 0.014
Definition of abbreviations: ARDS = acute respiratory distress syndrome; ICU = intensive care unit; IQR = interquartile range. Percentages are calculated on nonmissing data. Bold values indicate statistical significance. *Patients who received initially noninvasive ventilation but needed subsequent intubation were included in the invasive mechanical ventilation group. † Patients could have more than one pulmonary complication.
Cilloniz, Albert, Liapikou, et al.: Macrolide-Resistance Streptococcus pneumoniae Pneumonia
1269
ORIGINAL ARTICLE for multiple comparisons. Fifth, because we only encountered three patients with macrolide-resistant disease who were treated with macrolide monotherapy we cannot comment on the use of this regimen in this setting, although macrolide monotherapy would not be an appropriate choice for treating hospitalized patients.
Table 5. Antibiotic Regimens Macrolide Sensitive (n = 504) Monotherapy, n (%) Quinolone b-Lactam Macrolide Other Combination, n (%) b-Lactam 1 macrolide b-Lactam 1 quinolone Macrolide 1 quinolone Other Guideline-consistent regimens, n (%) Guideline-inconsistent regimens, n (%)
95 53 35 4 3 405 229 114 8 54 346 154
Macrolide Resistant (n = 139)
(19) (11) (7) (1) (1) (81) (46) (23) (2) (11) (69) (31)
34 22 7 3 2 104 69 22 1 12 91 47
(25) (16) (5) (2) (1) (75) (50) (16) (1) (9) (66) (34)
Conclusions
We found no evidence suggesting that patients hospitalized with culture- and roentgenographically proven, invasive or noninvasive S. pneumoniae pneumonia were more severely ill at presentation or had worse outcomes if their organism was resistant versus sensitive to macrolide antibiotics and/or if they were treated with guideline-compliant versus noncompliant regimens. Although the prevalence of in vitro resistance to macrolide antibiotics is increasing, we found no evidence that this resistance worsened outcomes in patients hospitalized for S. pneumoniae pneumonia. n
Percentages are calculated on nonmissing data.
patients admitted to other types of hospitals in other countries. Second, although we found no difference in the outcomes in patients with macrolideresistant versus -sensitive organisms, we could only analyze 139 patients with macrolide-resistant organisms and this sample size may result in a large type II error. Our sample size, however, is larger than many previous studies of macrolideresistant S. pneumoniae reported in the literature (14, 19, 20, 22, 32–34). Third, we did not genotype the S. pneumoniae
isolates to determine their specific mechanism of macrolide resistance and our failure to find any differences in outcomes could have resulted from a high fraction of the macrolide-resistant organisms having the ermB gene. A Spanish study found that 89.9% of macrolide-resistant pneumococci carried the ermB gene (35) but Daneman and colleagues (36) found equal representation of the mefA and ermB genes in their series of macrolide failures. Fourth, there were no adjustments made
Author disclosures are available with the text of this article at www.atsjournals.org. Acknowledgment: The authors are indebted to all medical and nursing colleagues for their assistance and cooperation in this study.
Table 6. Outcomes according to Appropriateness of Treatment and Macrolide Sensitivity Guideline-Consistent Regimens (n = 437)
Bacteremia, n (%) Days of hospital stay, median (IQR) 30-d in-hospital mortality, n (%) ICU admission, n (%) Mechanical ventilation,x n (%) None Noninvasive Invasive Pulmonary complications,jj n (%) Multilobar infiltration Pleural effusion ARDS Acute renal failure, n (%) Shock, n (%)
Macrolide Sensitive (n = 346)
Macrolide Resistant (n = 91)
Total (n = 437)
217 (63) 8 (5–12)
46 (51) 9 (5–14)
263 (61) 8 (5–13)
P Value
Guideline-Inconsistent Regimens (n = 201) Macrolide Sensitive (n = 154)
0.039 0.81
81 (53) 6.5 (4–11)
Macrolide Resistant (n = 47)
Total (n = 201)
19 (41) 6 (4–13)
100 (51) 6 (4–12)
P Value†
P Value‡
P Value
P Value*
0.15 0.74
0.040 0.005
0.28 0.10
0.017 0.001
27 (8)
6 (7)
33 (8)
0.70
13 (8)
6 (13)
19 (9)
0.38
0.81
0.22
0.42
109 (32)
23 (25)
132 (30)
0.25 0.046
36 (24)
11 (23)
47 (24)
0.99 0.58
0.071
0.81
0.081 0.040
265 21 32 166
(83) (7) (10) (49)
71 0 11 28
(87) (0) (13) (31)
336 21 43 194
(84) (5) (11) (45)
0.48 0.017 0.38 0.002
119 1 18 50
(86) (1) (13) (32)
35 1 7 16
(81) (2) (16) (34)
154 2 25 66
(85) (1) (14) (33)
0.44 0.38 0.59 0.84
0.44 0.007 0.35 0.001
0.44 0.17 0.66 0.70
0.74 0.018 0.29 0.004
116 69 24 119 47
(34) (20) (7) (35) (14)
19 13 5 33 6
(21) (14) (6) (37) (7)
135 82 29 152 53
(31) (19) (7) (36) (12)
0.020 0.20 0.58 0.80 0.061
32 28 4 34 20
(21) (18) (3) (22) (13)
12 7 1 7 3
(26) (15) (2) (15) (7)
44 35 5 41 23
(22) (18) (3) (21) (12)
0.49 0.59 0.88 0.31 0.24
0.040 0.63 0.052 0.004 0.79
0.54 0.92 0.40 0.009 0.99
0.019 0.67 0.033
