Selective decontamination of the digestive tract (SDD) refers to the administration of prophylactic antibiotics to critically ill patients in the hopes of either preventing or treating airway or digestive tract colonization by organisms that could potentially cause an infection. The rationale behind this practice posits that elimination of selected microorganisms from the oropharynx and upper gastrointestinal (GI) tract will prevent respiratory or bloodstream infections in critically ill patients.
The question of whether to implement SDD on a regular basis is somewhat unique. SDD has been extensively investigated through numerous randomized trials and multiple meta-analyses, with the preponderance of data supporting its use as beneficial. Despite the large literature supporting its use, a consensus on the appropriateness of SDD is lacking among critical care practitioners worldwide, and, in fact, large-scale adoption of SDD has not occurred because of continued concerns about SDD inducing antibiotic resistance.
Details of Selective Decontamination of the Digestive Tract
The prevention of aerodigestive tract colonization with pathogenic bacteria is the goal of SDD. Theoretically, selectively decreasing the bacterial populations of the upper digestive tract and airway of critically ill patients should decrease the risk of developing ventilator-associated pneumonia. As such, SDD protocols aim to selectively limit the presence of potentially harmful bacteria without adversely impacting the overall microbiome of the patient or the intensive care unit (ICU). Antibiotics used in SDD plans are therefore chosen to treat two different groups of microbes: endogenous bacteria already present in patients that could become pathogenic (such as Staphylococcus aureus or Streptococcus pneumoniae ), and gram-negative organisms that may secondarily colonize a patient during acute illness. Therefore SDD typically involves the administration of (1) a short intravenous course of a broad-spectrum cephalosporin aimed at treating existing, potentially pathogenic organisms and (2) ongoing enteral administration of nonabsorbable agents targeted toward gram-negative bacteria.
SDD should be contrasted with selective oral decontamination, or SOD, which treats only the oral cavity. Although SOD is frequently a component of broader SDD treatment plans, confusion may arise in interpreting study results because authors may varyingly consider SDD and SOD to be identical or separate interventions. For the purpose of clarity, this chapter distinguishes between the two whenever possible.
Representative examples of treatment strategies for both SDD and SOD are listed in Table 46-1 . SDD contains three components: (a) third-generation cephalosporins are dosed intravenously during the first 4 to 5 days of ICU admission, (b) nonabsorbable enteral antibiotics are given in a liquid form through a nasoenteric tube, and (c) pastes or gels are given to the oropharynx. The most commonly used enteral and oral agents are amphotericin, colistin, and tobramycin, although several other agents have also been studied. Although the term SDD is universally used, in actuality, it is a misnomer because there are multiple components to successful treatment regiments, including elements that are not directed at the digestive system per se.
| Oral | Enteral | Parenteral | |
|---|---|---|---|
| SDD | Paste of amphotericin, colistin, tobramycin, 2% each, applied to oropharynx q6 hr for duration of ICU stay | Amphotericin (500 mg), colistin (100 mg), tobramycin (80 mg) combined in 10 mL liquid suspension, administered via nasogastric tube q6 hr for duration of ICU stay | Cefotaxime 1 g q6 hr or ceftriaxone 2 g q24 hr for 4 days on ICU admission |
| SOD | Paste of amphotericin, colistin, tobramycin, 2% each, applied to oropharynx q6 hr for duration of ICU stay | –– | –– |
In contrast to SDD, SOD omits parenteral and enteral treatments and uses only oral pastes.
Evidence on the Efficacy of Selective Decontamination of the Digestive Tract
SDD has been extensively studied. Investigations include numerous randomized trials and meta-analyses with results that generally indicate benefit in ICU patients. An initial publication from the Netherlands in the early 1980s found that SDD significantly reduced both secondary colonization with pathogenic gram-negative organisms and associated infections in patients with severe trauma. Serial culture data documented decreased airway and GI colonization with pathogenic bacteria, and infection rates fell drastically in the SDD group (16% vs. 81%).
These initial findings prompted a large number of subsequent evaluations. Indeed, SDD is unique in critical illness trials related to the sheer size of the data pool that addresses it. To date, various iterations of SDD or SOD have been analyzed in more than 50 randomized controlled trials ( Table 46-2 ). As a group, the data almost uniformly demonstrate a significant reduction in infectious complications in patients receiving SDD/SOD treatment, although impact on mortality varies widely among studies. For example, a randomized trial in Dutch ICUs of nearly 1000 patients found that a typical SDD regimen significantly lowered rates of colonization by resistant organisms (16% vs. 26%) and was associated with a decreased ICU mortality (15% vs. 23%). Similarly, a subsequent, even larger Dutch trial (nearly 6000 patients) found that infectious complications were reduced in SDD and SOD treatment groups and although crude mortality rates did not differ from the control group, mortality also decreased modestly after adjustment for varying patient characteristics in experimental arms.
| Year | Author | No. of Subjects (Treatment/Control) | Patient Population | Treatment | Control | Outcomes |
|---|---|---|---|---|---|---|
| 1984 | Stoutenbeek | 181 (122/59) | Trauma | Amphotericin B (AB), polymyxin E (PE), tobramycin (T) | Controls were historical; nonrandomized trial | Infection rate
|
| 1987 | Unertl | 39 (19/20) | Mixed | AB, PE, gentamicin | Standard care | Respiratory infections
|
| 1988 | Kerver | 96 (49/47) | Mixed | AB, PE, T, IV cefotaxime | Placebo | Infection
|
| 1989 | Ulrich | 100 (48/52) | Mixed | AB, PE, norfloxacin, IV trimethoprim | Standard care | Respiratory infection
|
| 1990 | Flaherty | 107 (51/56) | Cardiac surgery | PE, gentamicin, nystatin | Sucralfate | Infection
|
| 1990 | Rodriguez-Roldan | 28 (15/13) | Mixed | AB, PE, T | Placebo | Tracheobronchitis
|
| 1990 | Tetteroo | 114 (56/58) | Esophageal surgery | AB, PE, T, IV cefotaxime | Standard care | Total infections
|
| 1991 | Aerdts | 56 (17/18 + 21) | Mixed | AB, PE, norfloxacin, iv cefotaxime | Standard care | Lower respiratory tract infections
|
| 1991 | Blair | 256 (126/130) | Mixed | AB, PE, T, IV cefotaxime | Placebo | Infection
|
| 1991 | Pugin | 79 (38/41) | Trauma | SOD only: PE, neomycin, vancomycin | Placebo | Pneumonia
|
| 1991 | Zobel | 50 (25/25) | Pediatric | AB, PE, gentamicin, IV cefotaxime | Standard care | Infection
|
| 1992 | Cerra | 46 (23/23) | Surgical | Norfloxacin, nystatin | Placebo | Total infections
|
| 1992 | Cockerill | 150 (75/75) | Mixed | PE, gentamicin, nystatin | Placebo | Total infections
|
| 1992 | Gastinne | 445 (220/225) | Mixed | AB, PE, T | Placebo | No change in pneumonia or mortality |
| 1992 | Hammond | 239 (114/125) | Mixed | AB, PE, T, IV cefotaxime | Placebo | No change in infection rate or mortality |
| 1992 | Rocha | 101 (47/54) | Mixed | AB, PE, T, IV cefotaxime | Placebo | Overall infection
|
| 1992 | Winter | 183 (91/92) 84 historic | Medical | AB, PE, T, IV ceftazidime | Standard care | Total infections
|
| 1993 | Korinek | 123 (63/60) | Neurosurgical | AB, PE, T, vancomycin added to oral solution | Placebo | Pneumonia
|
| 1993 | Rolando | Group 1: 21 Group 2: 21 Group 3: 28 Group 4: 31 | Hepatic failure | 1: IV cefuroxime 2: AB, PE, T, IV cefuroxime 3: AB, PE, T, IV cefuroxime | 4: Standard care | Total infections (Group 3 vs. Group 4)
|
| 1994 | Bion | 59 (27/32) | Liver transplant | AB, PE, T, IV cefotaxime, IV ampicillin | Nystatin, IV cefotaxime, IV ampicillin | Infections
No change in multiorgan dysfunction |
| 1994 | Ferrer | 80 (39/41) | Mixed | AB, PE, T, IV cefotaxime | Placebo, IV cefotaxime | No change in infection rate, pneumonia, or mortality |
| 1994 | Laggner | 67 (33/34) | Mixed | Oral gentamicin only | Placebo | No change in pneumonia or mortality |
| 1995 | Luiten | 102 (50/52) | Pancreatitis | AB, PE, enteral norfloxacin | Standard care | Mortality
|
| 1995 | Wiener | 61 (30/31) | Mixed | AB, PE, gentamicin | Placebo | No change in infection rate, pneumonia, or mortality |
| 1996 | Arnow | 69 (34/35) | Liver transplant | AB, PE, T, IV cefotaxime, IV ampicillin | IV cefotaxime, IV ampicillin | Aerobic gram-negative infections
|
| 1996 | Quinio | 148 (76/72) | Trauma | AB, PE, gentamicin | Placebo | Total infections
|
| 1996 | Rolando | 108 (47/61) | Hepatic failure | AB, PE, T, IV ceftazidime, flucloxacillin | AB, IV ceftazidime, flucloxacillin | No change in infection rate or mortality |
| 1997 | Abele-Horn | 88 (58/30) | Surgical | SOD only: AB, PE, T | Placebo | Primary pneumonia
|
| 1997 | Lingnau | 313 Group 1: 83 Group 2: 82 Control: 148 | Trauma | Group 1: AB, PE, T, IV ciprofloxacin Group 2: AB, PE, IV ciprofloxacin | Placebo, IV ciprofloxacin | No change in rates of pneumonia, sepsis, organ dysfunction, or mortality |
| 1997 | Schardey | 205 (102/103) | Surgical | AB, polymyxin B, T, oral vancomycin, IV cefotaxime | Placebo | Anastomotic leak
|
| 1997 | Verwaest | 615 Group 1: 195 Group 2: 200 Control: 220 | Mixed | Group 1: AB, ofloxacin enteral and IV Group 2: AB, PE, T, IV cefotaxime | Standard care | Group 1 vs. Group 2: Infection
Resistant organisms
|
| 1998 | Ruza | 226 (116/110) | Pediatric | PE, T, nystatin | Standard care | No change in infection rate or mortality |
| 1998 | Sanchez Garcia | 271 (131/140) | Trauma | AB, PE, oral and enteral gentamicin, IV ceftriaxone | Placebo | VAP
|
| 2001 | Barret | 23 (11/12) | Pediatric burn | AB, PE, T | Placebo | No difference in pneumonia or sepsis |
| 2001 | Begmans | 226 87 Control in same ICU: 78 (Group A) Control different setting: 61 (Group B) | Mixed | SOD only: PE, gentamicin, vancomycin | Placebo | VAP
|
| 2002 | Bouter | 51 (24/27) | Cardiac bypass | PE, neomycin | Placebo | Aerobic gram-negative carriage
|
| 2002 | Hellinger | 80 (37/43) | Liver transplant | PE, nystatin, gentamicin | Nystatin | No change in infection rate or mortality |
| 2002 | Krueger | 527 (265/262) | Surgical | PE, gentamicin, IV ciprofloxacin | Placebo | Total infection
|
| 2002 | Pneumatikos | 61 (30/31) | Trauma | AB, PE, T (subglottic decontamination only) | Placebo | Pneumonia
|
| 2002 | Rayes | 95 Group 1: 32 Group 2: 31 Control: 32 | Liver transplant | Group 1: AB, PE, T Group 2: Enteral fiber, Lactobacillus plantarum 299 | Placebo | Group 1 vs. Group 2: Infection
No change in infections No change in LOS for all comparisons |
| 2002 | Zwaveling | 55 (26/29) | Liver transplant | AB, PE, T | Placebo | No change in rate of infection |
| 2003 | de Jonge | 934 (466/468) | Surgical | AB, PE, T, IV cefotaxime | Standard care | ICU mortality
|
| 2005 | Camus | 515 Group 1: 130 Group 2: 130 Group 3: 129 Control: 126 | Mixed | Group 1: PE, T Group 2: Nasal mupirocin, chlorhexidine wash Group 3: Both treatments | Placebo | Group 3 vs. Control: Infections
|
| 2005 | de la Cal | 107 (53/54) | Burn | AB, PE, T | Placebo | Mortality
|
| 2006 | Gosney | 203 (103/100) | Stroke | SOD only: AB, PE | Placebo | Pneumonia
|
| 2006 | Koeman | 385 Group 1: 127 Group 2: 128 Control: 130 | Mixed | SOD only: Group 1: Chlorhexidine Group 2: Chlorhexidine, PE | Placebo | Pneumonia Group 1:
|
| 2007 | Stoutenbeek | 401 (200/201) | Trauma | AB, PE, T, IV cefotaxime | Standard care | Respiratory infection
|
| 2008 | Farran | 91 (40/51) | Surgical | Erythromycin, gentamicin, nystatin | Placebo | No change in anastomotic leak rate, pneumonia, or mortality |
| 2009 | de Smet | 6299 Group 1: 1904 Group 2: 2405 Control: 1990 | Mixed | Group 1: SOD only, AB, PE, T Group 2: SDD, AB, PE, T, IV cefotaxime | Standard care | Gram-negative infections SOD:
SOD:
|
| 2011 | Roos | 289 (143/146) | Surgical | AB, T, polymyxin B | Placebo | Infectious complications
|
| 2011 | de Smet (post hoc analysis from de Smet, 2009) | 5927 Group 1: 1904 Group 2: 2034 Control: 1989 | Mixed | Group 1: SOD only, AB, PE, T Group 2: SDD, AB, PE, T, IV cefotaxime | Standard care | Bacteremia SOD:
SOD:
SOD:
|
| 2012 | Melsen (post hoc analysis from de Smet, 2009) | 5927 Surgical: Group 1: 866 Group 2: 923 Control: 973 Medical: Group 1: 1038 Group 2: 1111 Control: 1016 | Mixed | Group 1: SOD only, AB, PE, T Group 2: SDD, AB, PE, T, IV cefotaxime | Standard care | Mortality in nonsurgical patients SOD:
Mortality in surgical patients SOD: no change in mortality SDD: no change in mortality |
Related posts:
What Lessons Have Intensivists Learned During the Evidence-Based Medicine Era?
What Is the Role of Invasive Hemodynamic Monitoring in Critical Care?
Should Fever Be Treated?
Are Computerized Algorithms Useful in Managing the Critically Ill Patient?
Inhaled Vasodilators in ARDS: Do They Make a Difference?
How Can We Monitor the Microcirculation in Sepsis? Does It Improve Outcome?
Full access? Get Clinical Tree
