DOI: 10.1128/CMR.00002-14
Article Figures & Data
Figures
- FIG 1
Diagnostic hierarchy from positive blood cultures to community-onset bloodstream infection. BSI, bloodstream infection; HO, hospital onset; HCA, health care associated; CA, community associated. The overall triangular area indicates all positive blood cultures, which may represent contamination (first level), transient bacteremia/fungemia (second level), or “true” BSI (third level). Bloodstream infections are further classified into the three mutually exclusive categories of hospital-associated, health care-associated community-onset, and community-associated community-onset BSIs.
Tables
- TABLE 1
Definitions of bloodstream infection and associated entities
Entity Definition Contamination of blood cultures Blood cultures are positive for growth due to organisms that were not present in the bloodstream Bacteremia Presence of viable bacteria in the blood; blood cultures positive for bacterial growth where contamination has been ruled out Fungemia Presence of viable fungi in the blood; blood cultures positive for fungal growth where contamination has been ruled out Transient bacteremia/fungemia Brief episode of bacteremia/fungemia that is not associated with infection Bloodstream infection Bacteremia/fungemia that is associated with infection Hospital-onset bloodstream infection Bloodstream infection that is first identified (culture drawn) ≥48 h after hospital admission and within 48 h following hospital discharge Community-onset bloodstream infection Bloodstream infection occurring in an outpatient or first identified (culture drawn) <48 h following admission to hospital Health care-associated community-onset bloodstream infection Community-onset bloodstream infection associated with significant prior health care exposure (as evidenced by recent hospitalization, specialized in-home medical services, care in a hospital-based clinic or hemodialysis unit, or residence in a nursing home) Community-associated community-onset bloodstream infection Community-onset bloodstream infection not fulfilling criteria for health care-associated infection Polymicrobial bloodstream infection Episode of bloodstream infection associated with two or more different organisms isolated within 48 h of each other - TABLE 2
Studies comparing health care-associated (HCA) and community-associated (CA) community-onset BSIs
First author (reference) Setting/population HCA BSI definition Patient differences (vs CA BSI) Microbiology (vs CA BSI) Outcome difference (vs CA BSI) Friedman (65) Adults admitted to three hospitals in North Carolina (143 with CA BSI and 186 with HCA BSI) Community-onset BSI with any specialized therapy in the home, recent attendance at a hospital, hemodialysis, or chemotherapy clinic, recent hospitalization, or residence in a nursing home More likely to have cancer and renal failure with HCA BSI, and less likely to have HIV infection Urinary tract infection more common in CA BSI; Staphylococcus aureus most common in HCA BSI and Escherichia coli and Streptococcus pneumoniae most common in CA BSI; MRSA much more common in HCA BSI 3- to 6-month mortality higher (29% versus 16%; P = 0.019) for HCA BSI than for CA BSI Lenz (66) Adult residents of Calgary, Canada (3,088 with CA BSI and 2,492 with HCA BSI) Minor modification of criteria of Friedman et al. (60, 65) Older, more comorbid illness with HCA BSI Different distribution of pathogens and higher rates of resistant organisms, including MRSA; more polymicrobial infections with HCA BSI Longer length of stay and higher 28-day case fatality rate (18% versus 10%; P < 0.001) with HCA BSI Al-Hasan (67) Gram-negative BSI in residents of Olmsted County, MN (306 with HCA BSI and 427 with CA BSI) Per criteria of Friedman et al. (65) Patients with HCA BSI were older Different distribution of infection foci and pathogens; higher rates of resistance with HCA BSI Higher 28-day case fatality rate (15% versus 4%; P < 0.001) with HCA BSI Son (50) Patients admitted to nine university hospitals in Korea (380 with CA BSI and 206 with HCA BSI) Per criteria of Friedman et al. (65) Patients with HCA BSI more likely to be male and to have comorbidities and immune-suppressant therapy Different distribution of infection foci and pathogens; higher rates of resistance with HCA BSI Higher 30-day case fatality rate (18% versus 10%; P = 0.007) with HCA BSI Kollef (68) Adults admitted to seven hospitals in the United States (728 [64%] with HCA BSI and 415 with CA BSI) Recent hospitalization, immune suppression, hemodialysis, or nursing home residence Patients with HCA BSI were older and more likely to be male and to have comorbidities and a higher severity of illness Different distribution of infection pathogens; higher rates of primary BSI and resistance with HCA BSI Higher hospital case fatality (14% versus 4%; P < 0.001) with HCA BSI Evans (69) Adults with spinal cord injury admitted to two hospitals in the United States (110 with HCA BSI and 36 with CA BSI) Per criteria of Friedman et al. (65) Patients with HCA BSI were older and more likely to have comorbidities Trend for higher rates of resistance with HCA BSI No difference in hospital or 30-day mortality rates Valles (70) Adults admitted to three teaching hospitals in Spain (581 with CA BSI and 281 with HCA BSI) Per criteria of Friedman et al. (65) Patients with HCA BSI were older and more likely to have comorbidities Different distribution of pathogens and higher rates of resistance, including MRSA; more polymicrobial infections with HCA BSI Higher case fatality rate (28% versus 10%; P < 0.001) with HCA BSI - TABLE 3
Population-based studies investigating the overall burden of disease associated with community-onset bloodstream infection
First author (reference) Setting, yr Annual incidence Case fatality rate (%) Mortality per 100,000 population Filice (1) Charleston County, SC (population 250,000), 1974–1976 43 per 100,000 Not reported Not reported Uslan (2) Olmsted County, MN (population 124,277), 2003–2005 154 per 100,000 (70 per 100,000 were health care associated, and 84 were community acquired) Not reported Not reported Sogaard (3) Northern Denmark (population 500,000), 1992–2006 112 per 100,000 (79 per 100,00 were CA, and 34 were HCA) during 2002 to 2006 17 (15 for CA BSI and 22 for HCA BSI) during 2002 and 2006 19 (12 for CA BSI and 7 for HCA BSI) during 2002 to 2006 Laupland (4) Calgary area, Alberta, Canada (population 1 million), 2000–2004 81.6 per 100,000 13 (in hospital) 11 (in hospital) Laupland (5) Victoria area, British Columbia, Canada (population 358,000), 1998–2005 101.2 per 100,000 12.6 (in hospital) 12.8 (in hospital) - TABLE 4
Contemporary population-based studies investigating the overall burden of Staphylococcus aureus community-onset bloodstream infection
First author (reference) Population, yr MSSA incidence per 100,000/yr MRSA incidence per 100,000/yr Total incidence per 100,000/yr Mortality (%) Morin (63) Four areas in Connecticut, 1998 14.5 2.5 17 10 for MSSA, 14 for MRSA, 11 overall (case fatality rates) Lyytikainen (82) Finland, 1995–2001 7 <0.1 7 13 (28-day case fatality rate) Collignon (83) Australia, 1999–2002 15 2 17 18 (28-day case fatality rate) Huggan (84) Canterbury, New Zealand, 1998–2006 14 <0.3 14 Not reported Laupland (85) and Tom (86) Multinational, 2000–2008 30-day case fatality rates of 20.2 for MSSA and 22.3 for MRSA, mortality rates of 3.4 for MSSA and 0.3 for MRSA per 100,000 per year Canberra, Australia 14.5 2.0 16.5 Calgary, Canada 15.3 2.4 16.5 Victoria, Canada 15.1 2.4 17.4 Sherbrooke, Canada 13.0 1.8 14.8 Northern Denmark 16.2 0 16.2 Copenhagen County, Denmark 14.7 0.2 14.9 Copenhagen City, Denmark 16.5 0.2 16.7 Finland 11.9 0.2 12.1 Western Sweden 18.1 0 18.2 Overall 15.0 1.0 16.0 - TABLE 5
Contemporary studies of invasive disease and/or bloodstream infection due to Streptococcus pneumoniae in nonselected populations
First author (reference) Population, yr Annual invasive disease incidence per 100,000 BSI rate or proportion (%) of isolates in blood Rudnick (108) Toronto, Canada, 1995–2011 9 Not reported Feemster (109) Philadelphia, PA, 2005–2008 Not reported 12.3 per 100,000 Helferty (110) Northern Canada, 1999–2010 25.8 Not reported Ingels (111) Denmark, 2000–2010 17.7 (2008–2010) Not reported Rosen (112) Eight U.S. regions, 1998–2009 13.5 (overall; range of 11.2 to 18 between regions during 2009) Not reported Harboe (113) Denmark, 1938–2007 Approximately 20 (1998–2007) 81 Weatherholtz (114) Navajo Nation, USA, 1995–2006 51 (2004–2006) Not reported Motlova (115) Czech Republic, 2000–2006 2.3–4.3 Not reported Hsieh (116) Taiwan, 2007 2.6 ≥94 Kellner (117) Calgary, Canada, 1998–2007 11.2 (2003–2007) 89 Baggett (118) Two areas in rural Thailand, 2005–2007 Not reported 3.7 per 100,000 (Sa Kaeo Province) and 7.6 per 100,000 (Nakhon Phanom Province) Bruce (119) Seven circumpolar regions, 1999–2005 21.6 for Canada (2003–2005), 19.8 for Greenland (2000–2005), 16.2 for Norway (2000–2005), 15.8 for Alaska (2001–2005), 14.6 for Iceland (2000–2005), 11.6 for Finland (2000–2005), 9.1 for Sweden (2003–2005) Not reported Heffernan (120) New Zealand, 1998–2005 12.4 93 Lacapa (121) Fort Apache Indian Reserve, USA, 1991–2006 87 (2001–2006) Not reported Klemets (122) Finland, 1995–2002 10.6 9.9 per 100,000 Einarsdottir (123) Iceland, 1975–2004 16.7 (1990–2004) Not reported Stephens (124) Atlanta, GA, 1994–2002 18 (2000–2002) 95 Andresen (125) Canberra, Australia, 1998–2000 15.2 Not reported
