Background Although long-term care facility (LTCF) residents are at increased risk for colonisation/infection with multidrug-resistant (MDR) organisms, few data are available on the prevalence of colonisation due to MDR-gram negative bacteria (GNB) in this setting, where infection control and preventive measures to reduce the spread of MDR bacteria are not well defined. The study investigated the prevalence and differences in the epidemiology of rectal colonisation with extended-spectrum beta-lactamase (ESBL) and carbapenemase-producing GNB among residents in LTCFs. Methods A multicenter point prevalence survey (PPS) was conducted in seven LTCFs in Italy. A rectal swab was collected to identify ESBL and carbapenemase-producing isolates, using rapid phenotypic methods and, subsequently, multiplex and single PCR. Clinical and epidemiological variables were assessed on the day of the PPS, including LTCF length of stay, hospitalization within three months and surgery and antimicrobial therapy within one month, functional and mental status, presence of medical devices and chronic skin lesions. Proportions and 95% confidence interval (CI95) were estimated accounting for cluster effect of the sample hierarchical structure (data collection site). The associations between colonisation and risk factors were estimated with a multilevel multivariate logistic regression model, with the data collection site as random component, assessing variables with p < 0.05 from bivariate analysis. Odds ratios (ORs) and CI95 were calculated to evaluate the associations. A p value < 0.05 was considered significant. Results A total of 453 residents were enrolled (74.6% females; mean age 83.7 years, SD 10.4 years). The majority of residents had urinary and/or faecal incontinence (81%), had spent more than one year in LTCF (78.4%), and had a low mental (70.2%) and functional (65.8%) status. 9.3% residents were administered penicillins/ß-lactamase inhibitors within the previous month, 4.2% were administered third-generation cephalosporins, 8.6% fluoroquinolones, and 9.7% cotrimoxazole (27.1% had received an antibiotic within the previous month). 16.6% had a urinary catheter, 15.7% had skin lesions, 14.6% had pressure sores, 9.5% had been hospitalised within 3 months, 3.8% had a vascular catheter, and 1.1% underwent surgery within the previous month. 7.7% (n = 35) was the prevalence of active infections on the PPS day. 21 residents (4.6%) received at least one antibiotic on the PPS day. Residents colonised with ESBL (88.8% Enterobacteriaceae and 11.2% non-fermenting GNB) and carbapenemase-producing GNB (77.8% Enterobacteriaceae and 22.2% non-fermenting GNB) were 39.5% (CI95, 32.5%-47%) and 4% (CI95, 2.8%-5.6%), respectively. Prevalence of colonisation at site level ranged from 14.3% to 57.1% for ESBL and from no detected cases to 9.5% for carbapenemase-producing GNB. Risk factors differed between residents colonised with ESBL and those with carbapenemase-producing strains. Male gender (OR, 2.2; CI95, 1.4-3.7; p = 0.002), previous exposure to third-generation cephalosporins (OR, 3.9; CI95, 1.3-12; p = 0.016) and bedridden status (OR, 1.7; CI95, 1-2.9; p = 0.04) were independently associated with ESBL colonisation, while previous hospitalization (OR, 4.1; CI95, 1.3-13.1; p = 0.02) was the only risk factor independently associated with carbapenemase-producing GNB colonisation. Conclusions The study confirms a high colonisation prevalence with ESBL-producing GNB among LTCFs residents and shows an alarming rate of residents colonised with carbapenemase-producing GNB. A national comprehensive effort is needed to promote and coordinate active surveillance of MDR-GNB in LTCFs to inform infection control teams. Differences in risk factors suggest that the spreading of ESBL and carbapenemase-producing GNB follows different routes. Infection control and preventive measures should be tailored on the type of antimicrobial-resistant GNB.
Introduzione Nonostante le strutture di lungodegenza siano gravate da un elevato rischio di colonizzazione/infezione da batteri multi-resistenti (MDR) agli antibiotici, i dati relativi alla prevalenza di colonizzazione da batteri Gram-negativi MDR sono esegui in questo specifico contesto, dove, peraltro, le misure di controllo e prevenzione delle infezioni atte a ridurre la diffusione di tali patogeni non sono ben definite. Lo studio si propone di valutare la prevalenza e le caratteristiche epidemiologiche della colonizzazione rettale da batteri Gram-negativi (GNB) produttori di beta-lattamasi a spettro-esteso (ESBL) e di carbapenemasi tra i degenti in sette lungodegenze della Provincia di Verona. Metodi Uno studio pilota, multicentrico, di prevalenza puntuale (PPS) è stato condotto in sette lungodegenze della Provincia di Verona. Un tampone rettale è stato raccolto per identificare isolati produttori di ESBL e carbapenemasi, utilizzando test rapidi fenotipici e, successivamente, genotipici (PCR). Nel giorno prescelto per l’esecuzione della PPS, sono state valutate variabili cliniche ed epidemiologiche, tra cui durata della degenza in LTCF, eventuale ospedalizzazione nei tre mesi precedenti, eventuale intervento chirurgico nel mese precedente, somministrazione di antibiotico-terapia nel mese precedente, grado di autonomia funzionale e capacità cognitive, presenza di dispositivi medici e lesioni cutanee croniche. Proporzioni ed intervalli di confidenza sono stati stimati considerando l’effetto cluster determinato dalla struttura gerarchica del campione (in base alla struttura in cui sono stati raccolti i dati). Le associazioni tra colonizzazione e fattori di rischio sono state stimate mediante un modello di regressione logistica multivariata multilivello, considerando la singola struttura come componente random. Sono state prese in considerazione le variabili caratterizzate da p < 0.05 nell’analisi bivariata. Le associazioni sono state valutate calcolando Odds Ratios (ORs) e CI95. p < 0.05 sono stati considerati significativi. Risultati Nello studio sono stati arruolati complessivamente 453 residenti nelle lungodegenze (74.6% di sesso femminile; età media, 83.7 anni [SD 10.4 anni]). La maggior parte dei soggetti reclutati era residente da più di un anno in LTCF (78.4%) ed era caratterizzato da incontinenza urinaria e/o fecale (81%) nonché da un considerevole deterioramento delle capacità cognitive (70.2%) e del grado di autonomia funzionale (65.8%). Complessivamente, era stato somministrato un antibiotico nel mese precedente al 27.1% dei residenti, in particolare nel 9.3% dei soggetti erano state somministrate penicilline associate ad inibitori delle ß-lattamasi, nel 4.2% cefalosporine di terza generazione, nell’8.6% fluorochinoloni e nel 9.7% cotrimossazolo. Il 16.6% dei soggetti reclutati aveva un catetere urinario, il 14.6% ulcere da decubito, il 15.7% lesioni cutanee; il 9.5% era stato ospedalizzato nei tre mesi precedenti, il 3.8% disponeva di un catetere vascolare e l’1.1% era stato sottoposto ad intervento chirurgico nel mese precedente. Durante il giorno della rilevazione, il 7.7% (n = 35) era affetto da infezione ed il 4.6% (n = 21) stava assumendo terapia antibiotica. I degenti colonizzati con ceppi produttori di ESBL (88.8% Enterobacteriaceae e 11.2% GNB non-fermentanti) e GNB produttori di carbapenemasi (77.8% Enterobacteriaceae e 22.2% GNB non-fermentanti) erano 39.5% (CI95, 32.5%-47%) e 4% (CI95, 2.8%-5.6%), rispettivamente. La prevalenza di colonizzazione a livello di singola struttura variava dal 14.3% al 57.1% per i ceppi produttori di ESBL e da 0 a 9.5% per i GNB produttori di carbapenemasi. I fattori di rischio significativamente associati alla colonizzazione da MDR-GNB differivano tra i residenti colonizzati con GNB produttori di ESBL e quelli colonizzati con ceppi produttori di carbapenemasi. L’appartenenza al genere maschile (OR, 2.2; CI95, 1.4-3.7; p = 0.002), una pregressa esposizione alle cefalosporine di terza generazione (OR, 3.9; CI95, 1.3-12; p = 0.016) e l’allettamento (OR, 1.7; CI95, 1-2.9; p = 0.04) erano indipendentemente associati con la colonizzazione da GNB produttori di ESBL, mentre la pregressa ospedalizzazione (OR, 4.1; CI95, 1.3- 13.1; p = 0.02) rappresentava l’unico fattore di rischio indipendentemente associato con la colonizzazione da ceppi produttori di carbapenemasi. Conclusioni Lo studio conferma un’elevata prevalenza di colonizzazione da GNB produttori di ESBL tra i residenti nelle LTCFs ed evidenzia un allarmante tasso di colonizzazione da GNB produttori di carbapenemasi. Da questi dati emerge la necessità urgente di promuovere e coordinare a livello nazionale un sistema di sorveglianza attiva di MDR-GNB in LTCF, che guidi le attività di prevenzione e controllo delle infezioni. Le differenze rilevate nell’associazione di specifici fattori di rischio suggeriscono che la diffusione di GNB produttori di ESBL e carbapenemasi segua percorsi differenti. Le misure di controllo e prevenzione delle infezioni dovrebbero, pertanto, essere mirate in funzione del tipo di GNB antibiotico-resistente.
A point prevalence survey of healthcare-associated infections, antimicrobial use and rectal colonisation with extended-spectrum beta-lactamase and carbapenemase-producing Gram-negative bacteria in seven long-term care facilities
MAZZAFERRI, Fulvia
2019
Abstract
Background Although long-term care facility (LTCF) residents are at increased risk for colonisation/infection with multidrug-resistant (MDR) organisms, few data are available on the prevalence of colonisation due to MDR-gram negative bacteria (GNB) in this setting, where infection control and preventive measures to reduce the spread of MDR bacteria are not well defined. The study investigated the prevalence and differences in the epidemiology of rectal colonisation with extended-spectrum beta-lactamase (ESBL) and carbapenemase-producing GNB among residents in LTCFs. Methods A multicenter point prevalence survey (PPS) was conducted in seven LTCFs in Italy. A rectal swab was collected to identify ESBL and carbapenemase-producing isolates, using rapid phenotypic methods and, subsequently, multiplex and single PCR. Clinical and epidemiological variables were assessed on the day of the PPS, including LTCF length of stay, hospitalization within three months and surgery and antimicrobial therapy within one month, functional and mental status, presence of medical devices and chronic skin lesions. Proportions and 95% confidence interval (CI95) were estimated accounting for cluster effect of the sample hierarchical structure (data collection site). The associations between colonisation and risk factors were estimated with a multilevel multivariate logistic regression model, with the data collection site as random component, assessing variables with p < 0.05 from bivariate analysis. Odds ratios (ORs) and CI95 were calculated to evaluate the associations. A p value < 0.05 was considered significant. Results A total of 453 residents were enrolled (74.6% females; mean age 83.7 years, SD 10.4 years). The majority of residents had urinary and/or faecal incontinence (81%), had spent more than one year in LTCF (78.4%), and had a low mental (70.2%) and functional (65.8%) status. 9.3% residents were administered penicillins/ß-lactamase inhibitors within the previous month, 4.2% were administered third-generation cephalosporins, 8.6% fluoroquinolones, and 9.7% cotrimoxazole (27.1% had received an antibiotic within the previous month). 16.6% had a urinary catheter, 15.7% had skin lesions, 14.6% had pressure sores, 9.5% had been hospitalised within 3 months, 3.8% had a vascular catheter, and 1.1% underwent surgery within the previous month. 7.7% (n = 35) was the prevalence of active infections on the PPS day. 21 residents (4.6%) received at least one antibiotic on the PPS day. Residents colonised with ESBL (88.8% Enterobacteriaceae and 11.2% non-fermenting GNB) and carbapenemase-producing GNB (77.8% Enterobacteriaceae and 22.2% non-fermenting GNB) were 39.5% (CI95, 32.5%-47%) and 4% (CI95, 2.8%-5.6%), respectively. Prevalence of colonisation at site level ranged from 14.3% to 57.1% for ESBL and from no detected cases to 9.5% for carbapenemase-producing GNB. Risk factors differed between residents colonised with ESBL and those with carbapenemase-producing strains. Male gender (OR, 2.2; CI95, 1.4-3.7; p = 0.002), previous exposure to third-generation cephalosporins (OR, 3.9; CI95, 1.3-12; p = 0.016) and bedridden status (OR, 1.7; CI95, 1-2.9; p = 0.04) were independently associated with ESBL colonisation, while previous hospitalization (OR, 4.1; CI95, 1.3-13.1; p = 0.02) was the only risk factor independently associated with carbapenemase-producing GNB colonisation. Conclusions The study confirms a high colonisation prevalence with ESBL-producing GNB among LTCFs residents and shows an alarming rate of residents colonised with carbapenemase-producing GNB. A national comprehensive effort is needed to promote and coordinate active surveillance of MDR-GNB in LTCFs to inform infection control teams. Differences in risk factors suggest that the spreading of ESBL and carbapenemase-producing GNB follows different routes. Infection control and preventive measures should be tailored on the type of antimicrobial-resistant GNB.File | Dimensione | Formato | |
---|---|---|---|
Tesi PhD Mazzaferri.docx
Open Access dal 21/05/2021
Dimensione
1.01 MB
Formato
Microsoft Word XML
|
1.01 MB | Microsoft Word XML | Visualizza/Apri |
I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.14242/114758
URN:NBN:IT:UNIVR-114758