Print version ISSN 1806-3713
J. bras. pneumol. vol.38 no.4 São Paulo July/Aug. 2012
Impact of bacteremia in a cohort of patients with pneumococcal pneumonia*
Ileana PalmaI; Ricardo MosqueraII; Carmen DemierIII; Carlos A VayIV; Angela FamigliettiV; Carlos M LunaVI
IChief Resident in Pulmonology. Pulmonology Section, University of Buenos Aires Hospital de Clínicas, Buenos Aires, Argentina
IIResident in Pulmonology. Pulmonology Section, University of Buenos Aires Hospital de Clínicas, Buenos Aires, Argentina
IIIStaff Member of the Bacteriology Section. Department of Biochemistry, University of Buenos Aires Hospital de Clínicas, Buenos Aires, Argentina
IVHead of the Bacteriology Section. Department of Biochemistry, University of Buenos Aires Hospital de Clínicas, Buenos Aires, Argentina
VFull Professor. University of Buenos Aires School of Pharmacy and Biochemistry, Buenos Aires, Argentina
VIFull Professor and Head of the Pulmonology Section. Department of Medicine, University of Buenos Aires Hospital de Clínicas, Buenos Aires, Argentina
OBJECTIVE: Bacteremia is the most common presentation of invasive disease in community-acquired pneumonia (CAP) due to Streptococcus pneumoniae. We investigated whether bacteremia in pneumococcal CAP worsens outcomes and whether it is related to pneumococcal vaccination (PV).
METHODS: Secondary analysis of a cohort of patients with pneumococcal CAP confirmed by blood culture, sputum culture, or urinary antigen testing. Demographic, clinical, radiographic, and biochemical data were collected, as were Acute Physiology and Chronic Health Evaluation II (APACHE II) and pneumonia severity index (PSI) scores, comorbidities, and PV history. We drew comparisons between patients with bacteremic pneumococcal CAP (BPP) and those with non-bacteremic pneumococcal CAP (NBPP).
RESULTS: Forty-seven patients had BPP, and 71 had NBPP (confirmed by sputum culture in 45 and by urinary antigen testing in 26); 107 had some indication for PV. None of the BPP patients had received PV, compared with 9 of the NBPP patients (p = 0.043). Among the BPP patients, the mean age was higher (76.4 ± 11.5 vs. 67.5 ± 20.9 years), as were APACHE II and PSI scores (16.4 ± 4.6 vs. 14.1 ± 6.5 and 129.5 ± 36 vs. 105.2 ± 45, respectively), as well as the rate of ICU admission for cardiopathy or chronic renal failure (42.5% vs. 22.5%), whereas hematocrit and plasma sodium levels were lower (35.7 ± 5.8 vs. 38.6 ± 6.7% and 133.9 ± 6.0 vs. 137.1 ± 5.5 mEq/L, respectively), although mortality was similar (29.8% vs 28.2%).
CONCLUSIONS: In this population at high risk for CAP due to S. pneumoniae, the PV rate was extremely low (8.4%). Although BPP patients were more severely ill, mortality was similar between the two groups. Because PV reduces the incidence of BPP, the vaccination rate in at-risk populations should be increased.
Keywords: Pneumococcal vaccines; Pneumonia, bacterial; Streptococcus pneumoniae; Pneumococcal infections; Mortality; Epidemiology.
Infection with Streptococcus pneumoniae is the most common cause of community-acquired pneumonia (CAP) worldwide. This pneumococcus is a common colonizer of the nasopharynx in many healthy individuals, especially children (the so-called carrier state) and, by contiguity with the upper airway, can cause diseases such as otitis, sinusitis, and pneumonia; less commonly, it can invade sterile body sites, such as the pleura, meninges, and blood, causing an invasive disease.(1) Bacteremia has been reported to occur in 10% of cases of CAP,(2-4) and there is evidence associating the presence of bacteremia with poor outcomes in patients with CAP.(5) In a study of CAP in the pre-antibiotic era, Tilghman et al. reported a high mortality rate (78%) in adults with bacteremic pneumococcal pneumonia (BPP), compared with 28% in those with non-bacteremic pneumococcal pneumonia (NBPP).(6) More recently, Austrian & Gold reported a 17% mortality rate in adults with BPP treated with penicillin.(7) The primary objective of the present study was to determine whether bacteremia in pneumococcal CAP was associated with predictors of poor outcomes (mortality, length of stay, complications, etc.) and whether it was related to vaccination history.
This was a secondary analysis of a prospective cohort of patients diagnosed with CAP and treated in the Hospital de Clínicas José de San Martín emergency room, clinical medicine ward, or ICU (or any combination of the three) between June of 1997 and May of 2001. The Hospital de Clínicas is the University of Buenos Aires hospital. The hospital is a referral center and a primary care center, having approximately 400 beds. Because a significant number of patients admitted to the hospital are retired social security beneficiaries, individuals over 65 years of age are predominant.
The primary objective of the present study was to determine the association between bacteremia and predictors of poor outcomes (mortality, length of stay, complications, etc.) in patients with CAP. To that end, the primary variable was mortality, chosen mainly because of its objectivity and its obvious clinical importance. Secondary variables included the following: presence of bacteremia; hospitalization; ICU admission; mortality, as predicted by scores (pneumonia severity index [PSI], mental Confusion, Urea, Respiratory rate, Blood pressure, and age > 65 years [CURB-65], and Acute Physiology and Chronic Health Evaluation II [APACHE II]); being older than 65 years; being a nursing home resident; comorbidities; and biochemical data useful in screening for the presence of comorbidities and in determining clinical severity. Our secondary objective was to determine the impact of bacteremia on the severity of the manifestations of CAP in terms of demographic variables, physical examination variables, biochemical variables, and severity rates, as well as to determine the relationship of bacteremia with pneumococcal vaccination history and the presence of various comorbidities.
The inclusion criteria were as follows: being over 18 years of age; having been diagnosed with CAP in accordance with the criteria defined by Fang et al. (i.e., presence of a new infiltrate on chest X-ray, together with one of the major criteria [fever > 38.0ºC, hypothermia < 35.0ºC, cough, and pleuritic pain] or two of the minor criteria [dyspnea, leukocytosis > 12,000 cells/mm3, altered level of consciousness, auscultatory signs of consolidation, and expectoration])(8); not having been hospitalized for more than 48 h prior to symptom onset; and presenting with positive blood culture for S. pneumoniae, positive sputum culture for S. pneumoniae, positive urinary antigen test results for S. pneumoniae, or any combination of the three.
The exclusion criteria were as follows: having been hospitalized in the last 30 days (or presenting with any other evidence suggestive of nosocomial pneumonia); having severe immunosuppression (e.g., having been under treatment with corticosteroids at a dose greater than 20 mg per day for more than 30 days); receiving chemotherapy for a lymphoproliferative disease or any other neoplastic disease; having neutropenia of any cause (< 1,000 cells/mm3); and being HIV-positive and having an AIDS-defining illness or a CD4+ lymphocyte count < 200 cells/mm3.
Clinical evaluation consisted of anamnesis and physical examination, the following data having been collected: demographic data (age, gender, and nursing home residence); data on flu and pneumococcal immunizations; data on clinical parameters at admission (temperature, blood pressure, HR, and RR, as well as level of consciousness, as determined by the Glasgow coma scale score); data on coexisting conditions (diabetes mellitus, chronic lung diseases such as COPD and asthma, chronic heart failure, chronic liver or kidney disease, neurological disease [dementia and stroke], HIV infection, and alcoholism); biochemical data (blood workup, urea, creatinine, glucose, ionography, and acid-base status); data on pneumonia severity (hospitalization and ICU admission), including APACHE II score,(9) CURB-65 score,(10) and PSI class(11); and data on 30-day mortality (regardless of the cause) and premature death, the latter having been defined as death occurring in the first 7 days.
We defined BPP as a diagnosis of pneumonia with one or more positive blood cultures for S. pneumoniae. In contrast, we defined NBPP as a diagnosis of pneumonia with positivity for S. pneumoniae as determined by sputum culture, an immunochromatographic test (Binax Now, S. pneumoniae urinary antigen test; Binax Inc, Portland, ME, USA), or both and negativity for S. pneumoniae as determined by blood cultures.
The statistical analysis was performed with the Statistical Package for the Social Sciences, version 15.0 for Windows (SPSS Inc., Chicago, IL, USA). We used the chi-square test or Fisher's exact test for categorical variables and the Student's t-test for continuous variables. In order to analyze discrete variables associated with mortality, we employed a predictive logistic regression model including the predictors that were found to be statistically significant by the chi-square test (p < 0.001). The level of statistical significance was set at p < 0.05.
Over a 4-year period, 798 patients diagnosed with CAP were hospitalized. Of those, 158 were excluded from the analysis because they did not meet all of the inclusion criteria or because they met one of the exclusion criteria. Of the remaining 640 patients, S. pneumoniae was found to be the etiologic agent in 118, who were therefore selected for inclusion in the study. Forty-seven met the criteria for BPP, and 71 met the criteria for NBPP. There were 23 outpatients. Of those, 5 were found to have bacteremia due to S. pneumoniae.
Among the BPP patients, the mean age was higher, the most common comorbidity being heart or kidney failure (Table 1). There were no differences between the two groups in terms of gender or any of the other comorbidities investigated. None of the BPP patients had received pneumococcal vaccination, whereas 9 (12.7%) of the NBPP patients had been vaccinated (p = 0.043). Of those 9, 7 were hospitalized (most being over 65 years of age and having severe pneumonia), and 2 (22.2%) died, whereas 32 (29.9%) of those who had not been vaccinated died (p = 0.624).
In addition to having hyponatremia, the BPP patients had significantly higher RRs and creatinine levels, as well as significantly lower hematocrit levels.
Although most of the BPP patients required hospitalization and ICU admission, there were no differences between the two groups in terms of the length of hospital stay. The BPP patients were more severely ill, as determined by APACHE II and PSI scores, the latter having shown a predominance of higher risk classes among those patients. The same was true for the CURB-65 score. However, we found no differences between the two groups in terms of mortality during hospitalization (Table 2).
Regarding the diagnostic methods, microbiological findings in sputum, urinary antigen testing, or both allowed us to identify the non-bacteremic group; it is of note that, although those methods were applied to the NBPP and BPP groups, the number of patients with microbiological findings in sputum was lower in the BPP group (Table 3).
Of the 118 patients, 34 (28.6%) died of pneumococcal pneumonia. Among those patients, being over 65 years of age, being a nursing home resident, having tachycardia and hypotension at admission, having a hematocrit level < 30%, and having a PaO2/FiO2 < 250 or oxygen saturation < 90% constituted variables that were significantly associated with higher mortality. The presence of chronic renal failure, chronic heart failure, or neurological disease was associated with higher mortality, as was the presence of more than one comorbidity. Other variables associated with a poor prognosis included ICU admission, CURB-65 > 2, and PSI class IV-V. However, when analyzing those characteristics in terms of the presence or absence of bacteremia, we found no statistically significant differences between the two groups (Table 4).
Logistic regression was performed to evaluate the impact of a number of variables on mortality in patients with pneumococcal pneumonia. As shown in Table 5, only three of the variables considered were found to be independently and significantly related to mortality in those patients: presence of more than one comorbidity (OR: 8.14); ICU admission (OR: 4.85); and presence of neurological disease (OR: 4.53).
Invasive pneumococcal disease causes high morbidity and mortality, especially in at-risk patients and older adults, as well as having a higher incidence in patients with lung disease, heart disease, and diabetes. The incidence of infection, especially bacteremia, empyema, and meningitis, in sterile sites varies by geographic region and ranges from 21 to 33 cases per 100,000 population.(12) In Argentina, a study conducted in the city of Tandil reported that the incidence of BPP was 2.8 cases/1,000 admissions or 17 cases/100,000 population per year.(13) Since the introduction of the pediatric 7-valent pneumococcal conjugate vaccine (PCV7) in 2002, the rate of invasive pneumococcal disease has decreased significantly in the United States, according to the Centers for Disease Control and Prevention. This reduction occurred not only among vaccinated children but also among unvaccinated adults and individuals over 65 years of age, through the so-called "herd effect".(14)
In the present study, we found certain differences between the characteristics of BPP and those of NBPP. We found that BPP was more common in older individuals than was NBPP, a finding that is in disagreement with those of previous studies.(15,16) We were unable to compare the incidence of BPP and NBPP in HIV-positive patients and intravenous drugs users with the incidence of BPP and NBPP in patients who did not have those characteristics because the incidence of HIV infection and intravenous drug use in our hospital population was zero. One third of our population had been diagnosed with COPD, a comorbidity that tended to be more common in NBPP patients (p = not significant). We found a predominance of BPP in patients with chronic heart disease or chronic renal failure, as has been reported, especially in elderly individuals who present with neutrophil dysfunction, decreased production of immunoglobulins, and malnutrition.(17)
Regarding clinical manifestations, the only significant difference between BPP patients and NBPP patients was that the former were more tachypneic. However, the BPP group was predominantly composed of elderly individuals, in whom tachypnea is more common,(18) Tachypnea is also a well-known indicator of severity found in many patients with bacteremia.
The BPP patients were found to have lower hematocrit levels and higher creatinine levels, as well as having hyponatremia, which is in agreement with previous reports by other authors.(19-22) We found a trend toward radiological evidence of pleural effusion, a finding that has previously been described in association with BPP.(19,20)
Bacteremia has traditionally been associated with poor outcomes in patients with CAP. In a study of CAP in the pre-antibiotic era, Tilghman et al. drew attention to the significant mortality in adults with BPP (78%), compared with 28% in those with NBPP.(6) More recently, Austrian & Gold reported a 17% mortality rate in adults with BPP treated with penicillin.(7)
In a recent study conducted in Canada and involving 1,154 patients with invasive pneumococcal disease, Marrie et al. found that being 18-40 years of age and receiving concurrent treatment with two antibiotics were associated with lower mortality.(23) The use of two antibiotics in combination for the treatment of severe forms of BPP is in line with the observations of Baddour et al.,(24) as well as with the Infectious Diseases Society of America/American Thoracic Society CAP management guidelines.(25) In addition, Marrie et al. observed that although premature death was not associated with comorbidities, it was associated with the presence of complications such as altered mental status, need for oxygen or mechanical ventilation, and cardiac arrest, all of which are indicators of the severity of the infection itself.(23) The mortality rate found by Marrie et al. in the aforementioned study was 14.1%, which is very similar to the 17% rate reported by Austrian & Gold in 1964, supporting the hypothesis of Austrian & Gold that, despite appropriate antibiotic therapy, invasive pneumococcal disease continues to have high mortality, which occurs during the first 3 days, in which the effect of antibiotics in reversing the fatal outcome seems to be nonexistent.(7,23)
Because a proportion of patients might die in the first days despite antibiotic use, it is clear that this group of patients would benefit from prevention. In Latin American countries, the pneumococcal polysaccharide vaccine is indicated for individuals over 65 years of age, those aged 2-64 years with comorbidities, and those cared for in nursing homes, as well as for immunocompromised individuals, according to different guidelines.(26,27) The 23-valent pneumococcal polysaccharide vaccine (PPV23) is well tolerated when administered for the first time but is seldom administered in the region, even to patients for whom it is clinically indicated.(28) In the present study, we confirmed the ability of PPV23 to prevent, in a significant manner, the development of BPP. However, 9 patients had NBPP despite having been vaccinated. A meta-analysis evaluating the effectiveness of PPV23 in function of patient age and risk level concluded that the vaccine is effective in preventing pneumococcal bacteremia in elderly individuals in general (65%) but less so in high-risk elderly individuals (20%), as well as showing little or no effectiveness in preventing pneumonia-the vaccine being even harmful (-16%)-in high-risk groups.(29) However, a recently published randomized controlled trial conducted in Japan and involving elderly nursing home residents (mean age > 84) demonstrated the usefulness of PPV23 in preventing NBPP.(30) Given that S. pneumoniae is the most common cause of pneumonia (25-50%), it would be very important to have a vaccine that can also protect against all forms of pneumococcal disease. Below the age of 2 years, the response to PPV23 is unsatisfactory, PCV7 being therefore indicated. It has been demonstrated that PCV7 can reduce the incidence of invasive and noninvasive pneumococcal disease, as well as reducing oropharyngeal colonization in children.(16) In this sense, there are expectations regarding the potential usefulness of conjugate vaccines that cover a larger number of serotypes and are close to being approved for use in adults.
In conclusion, BPP affects patients who are older and who are more severely ill. However, we found no significant differences between BPP patients and NBPP patients in terms of the length of hospital stay and mortality. Although it is more common in patients who are more severely ill, bacteremia does not determine severity; rather, it is an effect of disease severity. Although the pneumococcal polysaccharide vaccine prevents BPP, the levels of protection that the vaccine confers are not enough to prevent NBPP or colonization.
1. Musher DM. Streptococcus pneumoniae. In: Mandell GL, Douglas RG, Bennett JE, Dolin R, editors. Mandell, Douglas, and Bennett's Principles and Practice of Infectious Diseases. New York: Elsevier/Churchill Livingstone; 2005. p. 2623-42. [ Links ]
2. Plouffe JF, Breiman RF, Facklam RR. Bacteremia with Streptococcus pneumoniae. Implications for therapy and prevention. Franklin County Pneumonia Study Group. JAMA. 1996;275(3):194-8. [ Links ]
3. Balakrishnan I, Crook P, Morris R, Gillespie SH. Early predictors of mortality in pneumococcal bacteraemia. J Infect. 2000;40(3):256-61. [ Links ]
4. Afessa B, Greaves WL, Frederick WR. Pneumococcal bacteremia in adults: a 14-year experience in an inner-city university hospital. Clin Infect Dis. 1995;21(2):345-51. [ Links ]
5. Scott JA. The preventable burden of pneumococcal disease in the developing world. Vaccine. 2007;25(13):2398-405. [ Links ]
6. Tilghman RC, Finland M. Clinical significance of bacteremia in pneumococcic pneumonia. Arch Intern Med, 1937;59(4):602-19. [ Links ]
7. Austrian R, Gold J. Pneumococcal bacteremia with especial reference to bacteremic pneumococcal pneumonia. Ann Intern Med. 1964;60:759-76. [ Links ]
8. Fang GD, Fine M, Orloff J, Arisumi D, Yu VL, Kapoor W, et al. New and emerging etiologies for community-acquired pneumonia with implications for therapy. A prospective multicenter study of 359 cases. Medicine (Baltimore). 1990;69(5):307-16. [ Links ]
9. Knaus WA, Draper EA, Wagner DP, Zimmerman JE. APACHE II: a severity of disease classification system. Crit Care Med. 1985;13(10):818-29. [ Links ]
10. Lim WS, van der Eerden MM, Laing R, Boersma WG, Karalus N, Town GI, et al. Defining community acquired pneumonia severity on presentation to hospital: an international derivation and validation study. Thorax. 2003;58(5):377-82. [ Links ]
11. Fine MJ, Auble TE, Yealy DM, Hanusa BH, Weissfeld LA, Singer DE, et al. A prediction rule to identify low-risk patients with community-acquired pneumonia. N Engl J Med. 1997;336(4):243-50. [ Links ]
12. Robinson KA, Baughman W, Rothrock G, Barrett NL, Pass M, Lexau C, et al. Epidemiology of invasive Streptococcus pneumoniae infections in the United States, 1995-1998: Opportunities for prevention in the conjugate vaccine era. JAMA. 2001;285(13):1729-35. [ Links ]
13. Gentile JH, Sparo MD, Mercapide ME, Luna CM. Adult bacteremic pneumococcal pneumonia acquired in the community. A prospective study on 101 patients. Medicina (B Aires). 2003;63(1):9-14. [ Links ]
14. Whitney CG, Farley MM, Hadler J, Harrison LH, Bennett NM, Lynfield R, et al. Decline in invasive pneumococcal disease after the introduction of protein-polysaccharide conjugate vaccine. N Engl J Med. 2003;348(18):1737-46. [ Links ]
15. Marrie TJ, Low DE, De Carolis E; Canadian Community-Acquired Pneumonia Investigators. A comparison of bacteremic pneumococcal pneumonia with nonbacteremic community-acquired pneumonia of any etiology--results from a Canadian multicentre study. Can Respir J. 2003;10(7):368-74. [ Links ]
16. Musher DM, Rueda-Jaimes AM, Graviss EA, Rodriguez-Barradas MC. Effect of pneumococcal vaccination: a comparison of vaccination rates in patients with bacteremic and nonbacteremic pneumococcal pneumonia. Clin Infect Dis. 2006;43(8):1004-8. [ Links ]
17. Lipsky BA, Boyko EJ, Inui TS, Koepsell TD. Risk factors for acquiring pneumococcal infections. Arch Intern Med. 1986;146(11):2179-85. [ Links ]
18. Metlay JP, Schulz R, Li YH, Singer DE, Marrie TJ, Coley CM, et al. Influence of age on symptoms at presentation in patients with community-acquired pneumonia. Arch Intern Med. 1997;157(13):1453-9. [ Links ]
19. Musher DM, Alexandraki I, Graviss EA, Yanbeiy N, Eid A, Inderias LA, et al. Bacteremic and nonbacteremic pneumococcal pneumonia. A prospective study. Medicine (Baltimore). 2000;79(4):210-21. [ Links ]
20. Jover F, Cuadrado JM, Andreu L, Martínez S, Cañizares R, de la Tabla VO, et al. A comparative study of bacteremic and non-bacteremic pneumococcal pneumonia. Eur J Intern Med. 2008;19(1):15-21. [ Links ]
21. Brandenburg JA, Marrie TJ, Coley CM, Singer DE, Obrosky DS, Kapoor WN, et al. Clinical presentation, processes and outcomes of care for patients with pneumococcal pneumonia. J Gen Intern Med. 2000;15(9):638-46. [ Links ]
22. Torres JM, Cardenas O, Vasquez A, Schlossberg D. Streptococcus pneumoniae bacteremia in a community hospital. Chest. 1998;113(2):387-90. [ Links ]
23. Marrie TJ, Tyrrell GJ, Garg S, Vanderkooi OG. Factors predicting mortality in invasive pneumococcal disease in adults in Alberta. Medicine (Baltimore). 2011;90(3):171-9. [ Links ]
24. Baddour LM, Yu VL, Klugman KP, Feldman C, Ortqvist A, Rello J, et al. Combination antibiotic therapy lowers mortality among severely ill patients with pneumococcal bacteremia. Am J Respir Crit Care Med. 2004;170(4):440-4. [ Links ]
25. Mandell LA, Wunderink RG, Anzueto A, Bartlett JG, Campbell GD, Dean NC, et al. Infectious Diseases Society of America/American Thoracic Society consensus guidelines on the management of community-acquired pneumonia in adults. Clin Infect Dis. 2007;44 Suppl 2:S27-72. [ Links ]
26. Corrêa Rde A, Lundgren FL, Pereira-Silva JL, Frare e Silva RL, Cardoso AP, Lemos AC, et al. Brazilian guidelines for community-acquired pneumonia in immunocompetent adults - 2009. J Bras Pneumol. 2009;35(6):574-601. [ Links ]
27. Luna CM, Calmaggi A, Caberloto O, Gentile J, Valentini R, Ciruzzi J, et al. Pneumonia acquired in the community. Practical guide elaborated by a committee intersocieties [Article in Spanish]. Medicina (B Aires). 2003;63(4):319-43. [ Links ]
28. Donalisio MR, Rodrigues SM, Mendes ET, Krutman M. Adverse events after pneumococcal vaccination. J Bras Pneumol. 2007;33(1):51-6. [ Links ]
29. Melegaro A, Edmunds WJ. The 23-valent pneumococcal polysaccharide vaccine. Part I. Efficacy of PPV in the elderly: a comparison of meta-analyses. Eur J Epidemiol. 2004;19(4):353-63. [ Links ]
30. Maruyama T, Taguchi O, Niederman MS, Morser J, Kobayashi H, Kobayashi T, et al. Efficacy of 23-valent pneumococcal vaccine in preventing pneumonia and improving survival in nursing home residents: double blind, randomised and placebo controlled trial. BMJ. 2010;340:c1004. [ Links ]
Correspondence to: Submitted: 26 October 2011. * Study carried out in the Department of Medicine, Pulmonology Section, University of Buenos Aires Hospital de Clínicas, Buenos Aires, Argentina.
Carlos M. Luna
Arenales 25567, 1º A
CP 1425 Buenos Aires, Argentina
Tel. 54 11 5950-8931
Accepted, after review: 22 November 2011.
Financial support: None.
Submitted: 26 October 2011.
* Study carried out in the Department of Medicine, Pulmonology Section, University of Buenos Aires Hospital de Clínicas, Buenos Aires, Argentina.