Table 1.
Study Characteristics
| Study ID (Location) | Study Year | Design (No. of Centers) | Enrollment | Clinical Context | Definition of Pneumonia | Inclusion Criteria | Exclusion Criteria | Comparator Testsa | Reference Standard for Positive Diagnosis |
|---|---|---|---|---|---|---|---|---|---|
| Studies that assessed diagnostic accuracy | |||||||||
| Merrill 1973 [17] (Charlottesville, VA, US) | ND | ND (1) | ND | “Acute” pneumonia on admissionb | Acute change in health, fever >37.8°C, cough, new X-ray pulmonary infiltrates | Adults, “acute” pneumonia that required hospitalization, no previous antibiotic therapy | ND | Sputum antigen for Streptococcus pneumoniae by Quellung reaction | (1) Sputum culture only, (2) CRS (APR for cultures of multiple specimens) |
| Thorsteinsson 1975 [18] (Houston, TX, US) | ND | ND (1) | ND | “Acute” pneumonia on admissionb | Symptoms/signs of acute pneumonia, X-ray pulmonary infiltrates | “Acute” pneumonia that required hospitalization, no previous antibiotic therapy | ND | None | (1) Sputum culture only, (2) transtracheal aspirate culture only, (3) bronchial aspirate culture only, (4) CRS (APR by cultures of multiple specimens) |
| Rein 1978 [19] (Charlottesville, VA, US) | ND | Prospective (1) | Consecutive | CAP | Acute productive cough, new X-ray pulmonary infiltrates | CAP | ND | Sputum culture, mouse inoculation of sputum, sputum antigen for S. pneumoniae by Quellung reaction | (1) Sputum culture only, (2) CRS (APR for cultures of multiple specimens, mouse inoculation of sputum, sputum antigen for S. pneumoniae) |
| Boerner 1982 [20] (Durham, NC, US) | NDc | ND (1) | ND | CAP on admission | Symptoms/signs of acute RTI, X-ray pulmonary infiltrates or consolidations | CAP that required hospitalization | ND | None | CRS (APR by cultures of multiple specimens) |
| Dans 1984 [21] (Baltimore, MD, US) | 1971–1972, 1979–1980 | Retrospective (1) | Inconsecutive | CAP | Fever, X-ray pulmonary infiltrates, treating physicians’ clinical diagnosis | CAP that required hospitalization | CAP as the secondary diagnosis, incomplete data | None | Sputum culture only |
| BTS 1987 [22] (nationwide, UK) | Nov 1982–Dec 1983 | Prospective (25) | Inconsecutive | CAP | Acute symptoms, new segmental or lobar X-ray pulmonary infiltrates | Adults (15–74 y), CAP that required hospitalization | Pneumonia not the main reason of admission, pneumonia as the terminal event, pulmonary TB | None | CRS (APR by cultures of multiple specimens or sputum antigen for S. pneumoniae) |
| Zhang 1988 [23] (Shanghai, China) | Dec 1986 –Feb 1987 | ND (1) | Inconsecutive | CAP in ED | History/symptoms of acute LRTI, X-ray pulmonary infiltrations | CAP | No sputum collected | None | CRS (APR by cultures of multiple specimens, urine antigen for S. pneumoniae, or serology for S. pneumoniae) |
| Gleckman 1988 [24] (Worcester, MA, US) | Jan 1982–July 1987 | ND (1) | Consecutive | CAP with bacteremia on admission | Symptoms/signs of acute RTI, new X-ray pulmonary infiltrates | Adults, CAP with isolation of a bacterium from blood that required hospitalization | Any coexistent infection | None | (1) Sputum culture only, (2) blood culture only, (3) CRS (APR by cultures of sputum or blood) |
| Fine 1991 [25] (Pittsburgh, PA, US) | Jul 1986–Mar 1987 | Prospective (2) | Inconsecutive | CAP or HCAPd on admission | Symptoms/signs of LRTI, new X-ray pulmonary infiltrates | >16 y, CAP or HCAPd that required hospitalization | No sputum collected or missing results | None | CRS (APR by cultures of multiple specimens) |
| Bohte 1996 [26] (Leiden, Netherlands) | Jan 1991–Apr 1993 | Prospective (6) | Inconsecutive | CAP on admission | New X-ray pulmonary infiltrates | ≥18 y, CAP that required hospitalization | HCAP, hospitalization ≤1 wk, failures to obtain serologic tests, concomitant infection | None | CRS (APR for cultures of multiple specimens) |
| Roson 2000 [27] (Barcelona, Spain) | Feb 1995–May 1997 | Prospective (1) | Consecutive | CAP on admission | ≥1 signs/symptoms of LRTI, new X-ray pulmonary infiltrates | CAP that required hospitalization | Neutropenia, AIDS, transplantation, pneumonia of “unknown origin” | None | CRS (APR for cultures of multiple specimens and PCR of needle aspirate for S. pneumoniae) |
| Sato 2002 [28]e (Tokyo, Japan) | Jan 1997–Dec 2000 | Retrospective (1) | ND | CAP on admission | Acute signs/symptoms of LRTI, new X-ray pulmonary infiltrates | CAP that required hospitalization | Aspiration pneumonia, patients requiring ventilator, HCAPd | None | Sputum culture only |
| Butler 2003 [29] (Atlanta, GA, US) | Jan 1997–Mar 1998 | Retrospective (1) | Inconsecutive | CAP in ED | ≥1 signs/symptoms of LRTI, X-ray pulmonary infiltrates | ≥18 years, CAP that required hospitalization | Use of antimicrobials ≤7 d, no timely informed consent, HIV infection, anuria due to AKI/CKD, use of urinary catheter for >24 h, bleeding diathesis, abnormality/alteration of the upper respiratory tract | Urinary antigen for S. pneumoniae, sputum PCR | CRS (APR for cultures of multiple specimens) |
| Garcıa 2004 [30] (Barcelona, Spain) | Oct 1996–Apr 2002 | Prospective (1) | Consecutive | CAP in ED | Signs/symptoms of LRTI, new X-ray pulmonary infiltrates | >14 y, CAP | Neutropenia, HIV infection, TB, fungal infection, patients treated with immunosuppressive drugs, disease duration ≥ 2 wk | None | CRS (APR for cultures of multiple specimens) |
| Roson 2004 [31] (Barcelona, Spain) | Jun 2000–Apr 2002 | Prospective (1) | Consecutive | CAP | Acute respiratory illness, new X-ray pulmonary infiltrates | Adult, non–severely immunosuppressed, CAP | Neutropenia, AIDS, transplant recipients, pneumococcal vaccination ≤1 wk | Urinary antigen for S. pneumoniae | CRS (APR for cultures of multiple specimens) |
| Yang 2005 [32] (Baltimore, MA, US) | Oct 2001–May 2003 | Prospective (1) | Consecutive | CAP in ED | Acute signs/symptoms of LRTI, leukocytosis, new X-ray pulmonary infiltrates | >17 y, CAP, excess of sputum samples available, no missing data on reference standard | Failures to receive a reference standard | None | CRS (APR for cultures of multiple specimens, sputum or BAL fluid antigen for S. pneumoniae) |
| Miyashita 2008 [33] (Kurashiki, Japan) | Jan 2004–Jul 2007 | Prospective (1) | ND | CAP on admission | Signs/symptoms of LRTI, new X-ray pulmonary infiltrates | CAP that required hospitalization | HAP, HIV infection, use of immunosuppressive therapy or steroids, HAP | None | Sputum culture only |
| Anevlavis 2009 [34] (Athens and Alexandroupolis, Greece) | Jan 2002–Jun 2008 | Prospective (2) | Inconsecutive | CAP on admission | Signs/symptoms of LRTI, increased PMNs, X-ray pulmonary infiltrates | Selected “bacterial” CAP, no antimicrobial therapy <2 wk, same organism identified from both blood and sputum | ND | None | Both blood and sputum cultures positive |
| Ferre 2011 [35] (Barcelona, Spain) | Oct 2005–Nov 2007 | Retrospective (1) | Consecutive | Hospitalized CAP from ED on admission | Signs/symptoms of LRTI, X-ray pulmonary infiltrate | CAP that required hospitalization from ED | Pediatric or gynecology patients, cases requiring ICU care, empyema, immunosuppressed patients, HIV infection, patients on HD | None | CRS (APR for cultures of multiple specimens, urinary antigen for S. pneumoniae) |
| Fukushima 2013 [36] (nationwide, Japan) | Mar 2006–Mar 2007 | Prospective (14) | ND | CAP | ND | ≥16 y, CAP | ND | None | Sputum culture only |
| Akter 2014 [37] (Dhaka, Bangladesh) | Jul 2011–Jun 2012 | Prospective (1) | Consecutive | CAP | Fever, signs/symptoms of LRTI, new or progressing X-ray pulmonary infiltrates | >18 y, CAP | TB, BA, CHD, AKI/CKD, foreign body aspiration, current use of or recently completed antibiotic therapy | None | Sputum PCR for S. pneumoniae and H. influenzae |
| Fukuyama 2014 [38]e (Uruma, Japan) | Aug 2010–Jul 2012 | Prospective (1) | Consecutive | Hospitalized CAP or HCAPf from ED on admission | Signs/symptoms of LRTI, new X-ray pulmonary infiltrate | CAP or HCAPf that required hospitalization from ED | Nonpneumonia causes identified later through clinical follow–up | None | CRS (APR for cultures of multiple specimens, urinary antigen for S. pneumoniae) |
| Studies that assessed diagnostic yield | |||||||||
| Lim 1989 [39] (Adelaide, Australia) | Apr 1987–Mar 1988 | Prospective (1) | Consecutive | CAP on admission | Symptoms/signs of acute pneumonia, new X-ray pulmonary infiltrates | CAP that required hospitalization | Patients with immunosuppressive disorders, treated with immunosuppressive drugs, or with disorders that affects consciousness | Sputum culture, blood culture, viral culture | An operational diagnostic algorithm consisting of definitive and presumptive etiologies |
| van der Eerden 2005 [40] (Alkmaar, Netherlands) | Dec 1998–Nov 2000 | Prospective (1) | ND | CAP on admission | Symptoms/signs of acute pneumonia, new X-ray pulmonary consolidations | ≥18 y, CAP that required hospitalization | Severe immunosuppression, malignancy, pregnancy lactation, severe allergy to antibiotics obstruction pneumonia, ≤8 d after hospital discharge | Sputum culture, sputum and urine antigen for S. pneumoniae, urine antigen for Legionella pneumophila, and serological tests | An operational diagnostic algorithm consisting of definitive and presumptive etiologies |
Abbreviations: AKI, acute kidney injury; APR, any tests positive rule (ie, at least 1 positive result for any of the multiple reference standard tests performed was deemed as composite reference standard positive); BA, bronchial asthma; BAL, bronchoalveolar lavage; BTS, British Thoracic Society; CAP, community-acquired pneumonia; CHD, congenital heart disease; CKD, chronic kidney disease; CRS, composite reference standard; ED, emergency department; GA, Georgia; HD, hemodialysis; HAP, hospital-acquired pneumonia; HCAP, healthcare-associated pneumonia; HIV, human immunodeficiency virus; ICU, intensive care unit; LRTI, lower respiratory tract infection; MA, Massachusetts; MD, Maryland; NC, North Carolina; ND, no data; PA, Pennsylvania; PCR, polymerase chain reaction; PMN, polymorphonuclear; RTI, respiratory tract infection; TB, tuberculosis; TX, Texas; UK, United Kingdom; US, United States; VA, Virginia.
aOnly tests that were clearly defined and analyzed, the results of which were reported in comparison with the Gram stain results, were considered.
bNot specifically referred to as CAP.
cOne-year period.
dPatients from nursing home. These patients were included (see text).
e These studies also assessed diagnostic yield.
fAccording to the American Thoracic Society 2005 criteria. These patients were excluded from analysis.