Reiniging en desinfectie van ruimten Module 3 Evidence-tabellen
Evidence table for diagnostic accuracy studies
Subquestion 1 – Diagnostic performance of monitoring methods
Study reference |
Study characteristics |
Surface characteristics
|
Index test (test of interest) |
Reference test
|
Follow-up |
Outcome measures and effect size |
Comments |
---|---|---|---|---|---|---|---|
Amodio (2014) |
Type of study Cross-sectional study Setting and country Hospital, Italy Funding and conflicts of interest Not reported |
Inclusion criteria - Randomly selected, proportionally to the size of the hospital unit Exclusion criteria Surfaces Other important characteristics Type of surface |
Index test ATP (Adenosinetrifosfaat ) bioluminescence assay Cut-off point(s) Method
|
Reference test Microbiological culture: aerobic colony count (ACC) Cut-off point(s)* Method
|
Time between the index test and reference test Simultaneously Distance between sampling site for index test and reference test Adjacent Incomplete outcome data N=0 (0%) Reasons for incomplete outcome data Not applicable |
Correlation ATP values (log10RLU/cm2) vs. ACC (CFU/cm2): Pearson’s rho=0.54 R2=0.29
|
The authors conclude that ATP bioluminescence is a quick and sensitive test that can be a useful proxy of microbial contamination. No standardisation of contamination level. This may have resulted in reduced precision (difference in contamination level between adjacent surface areas). Area size sampled for reference test 25% of that sampled for index test. This lower sensitivity of the reference test may have resulted in false-negative results at low contamination levels and an overestimation of the correlation with the ATP bioluminescence method. Blinding of outcome assessment not reported for reference test. No incomplete data reported. It is considered unlikely that there were none. Outcome correlation is no measure of agreement. 95% CI of Pearson’s rho not reported. |
Trsan (2019) |
Type of study Cross-sectional study Setting and country Hospital, Slovenia Funding and conflicts of interest Not reported |
Inclusion criteria - Randomly selected Exclusion criteria Surfaces Category 2: n=150 Category 3: n=70 Category 4:n=208 Other important characteristics Type of surface |
Index test ATP bioluminescence assay Cut-off point(s)* Category 1: Category 2: Category 3: Category 4: Method
|
Reference test Microbiological culture Cut-off point(s)* Category 1: Category 2: Category 3: Category 4: Method
|
Time between the index test and reference test Simultaneously Distance between sampling site for index test and reference test Adjacent Incomplete outcome data N=0 (0%) Reasons for incomplete outcome data Not applicable |
Percentage of failures* ATP bioluminescence: 116/537 (21.6%; 95% CI: 18.3%-25.3%) Microbiological culture: 39/537 (7.3%; 95% CI: 5.4%-9.8%) Sensitivity Category 2: Category 3: Category 4: Specificity Category 2: Category 3: Category 4: |
The authors conclude that the ATP bioluminescence method is suitable and sensitive enough to be used in pharmacy cleanrooms as supplementary method, but that traditional microbiology remains the golden standard. Procedure for random selection of surface levels not reported. Considering adjacent sampling this is unlikely to have introduced bias. No standardisation of contamination level. This may have resulted in reduced precision (difference in contamination level between adjacent surface areas). Dependent on the required cleanliness of surfaces, thresholds for cleanliness differed for the index test and the reference test. This limits comparability of observed sensitivities and specificities for different surface categories. Blinding of outcome assessment not reported for reference test. No incomplete data reported. It is considered unlikely that there were none. |
* Pass = value below the specified cut-off point; fail = value equal to or above the specified cut-off point.
Evidence table for intervention studies
Subquestion 2 – Role of monitoring in infection control
Study reference |
Study characteristics |
Hospital/ward characteristics |
Intervention (I) |
Comparison / control (C) |
Follow-up |
Outcome measures and effect size |
Comments |
---|---|---|---|---|---|---|---|
Ziegler (2022) |
Type of study Multicenter, cluster-randomised controlled trial with pre-intervention baseline cohort. Setting and country Funding and conflicts of interest - Payed consultancy for company not manufacturing ATP bioluminescence or UV/F marker devices. - Receipt of equipment, ma- terials, drugs, medical writing, gifts, or other services from ATP biolumi- nescence and UV/F marker company |
Inclusion criteria Not reported Exclusion criteria Not reported N total ICUs at baseline ATP - UV/F: n=4 Important prognostic factors Not applicable (cross-over study) Daily and terminal cleaning |
Monitoring of cleaning performance with adenosine triphosphate (ATP) bioluminescence assay - Weekly Number of surfaces per month per ICU: Standard deviation 128 Range 83-801 *ATP vs. UV/F: p=0.002
|
Comparison 1 – UV/F Monitoring of cleaning performance with ultraviolet light inspection of fluorescent marker (UV/F) (= randomised) - Weekly Mean 353* Standard deviation 120 Range 83-658 *ATP vs. UV/F: p=0.002 Comparison 2 – VI Monitoring of cleaning performance with visual inspection - 10-30% of patient rooms |
Length of follow-up Both intervention periods: 6 months Washout: 2 months
Loss-to-follow-up: None reported
Incomplete outcome data: None reported
|
Hospital-acquired infection or colonisation with MDRO) (primary outcome) ATP vs. VI: IRR 0.89 (95% CI: 0.81-0.97) in favour of ATP UV/F vs. VI: IRR 1.10 (95% CI: 0.96-1.27) in favour of VI ATP vs. UV/F: IRR 0.88 (95% CI: 0.81-0.95) in favour of ATP Hospital-acquired infection with MDRO alone ATP vs. VI: IRR 0.92 (95% CI: 0.85-0.998) on favour of ATP UV/F vs. VI: IRR 1.02 (95% CI: 0.74-1.42) in favour of VI ATP vs. UV/F: IRR 0.93 (95% CI: 0.69-1.25) in favour of ATP
|
The authors conclude that intensive monitoring of ICU terminal room cleaning with an ATP bioluminescence assay is associated with a reduction of MDRO colonisation and infection. Open-label Power calculation: minimum detectable difference: 30% Number of surfaces monitored significantly higher for ATP period. Unclear whether this also pertains to feedback. Might have strengthened the effect of ATP. Comparison 2 is the comparison of interest for the clinical question of the current review. This concerns an uncontrolled before-after comparison without interrupted time series analysis. Concurrent changes in patient population or infection control practices not reported. Feedback of results was only performed for ATP and UV/F, not for VI. Trial not registered Potential conflict of interest
|