KLINIČNA UPORABNOST MOLEKULARNIH METOD DOKAZOVANJA POVZROČITELJEV INFEKCIJSKIH DRISK PRI ODRASLIH

Kotar, Tadeja (author), Lejko Zupanc, Tatjana (mentor), Pirš, Mateja (comentor)

Izhodišča in namen dela Z molekularnim diagnosticiranjem in s sindromskim naborom mikrobioloških preiskav (tj. s hkratnim pregledom blata na bakterije, viruse in parazite) se število drisk, ki jih vzročno ne uspemo pojasniti, pomembno zmanjša. Kljub temu ostaja še veliko odprtih vprašanj glede tolmačenja izvidov in njihove uporabnosti pri kliničnem delu. O razširjenosti (prevalenci) enteropatogenov v zdravi, brezsimptomni populaciji vemo malo, zato klinična vrednost teh najdb še ni jasna. V svetovni literaturi podatkov, ki bi neposredno primerjali rezultate v skupini bolnikov in rezultate v kontrolni skupini, ni veliko. Tudi poznavanje vrednosti pražnega cikla (angl. cycle of threshold, Ct) pri verižni reakciji s polimerazo v realnem času (angl. polymerase chain reaction, PCR) za posamezne mikroorganizme v vzorcih odraslih bolnikov z drisko in v vzorcih kontrolnih oseb ter opredelitev vloge Ct pri tolmačenju mikrobioloških izvidov za zdaj (še) nista zadovoljivi. Vzorec blata, odvzet v standardne posodice, je metoda izbire pri pridobivanju kužnine za dokaz enteričnih patogenov. Ker v določenih okoliščinah blata ne moremo dobiti že ob pregledu, bi kot alternativni način odvzema kužnine lahko uporabili bris rektuma. Večino raziskav o uporabnosti brisa rektuma so doslej opravili pri otrocih, medtem ko je podatkov o uporabnosti brisa rektuma za dokazovanje povzročitelja infekcijskih drisk pri odraslih za zdaj malo. Z raziskavo smo želeli opredeliti klinično uporabnost molekularnih diagnostičnih pristopov pri dokazovanju povzročiteljev infekcijske driske pri odraslih. Zanimalo nas je, katere potencialne povzročitelje driske dokažemo v vzorcih odraslih bolnikov z drisko in v vzorcih odraslih kontrolnih oseb brez črevesnih simptomov ter kako pogosto. Želeli smo tudi ugotoviti, ali se količina nukleinske kisline (DNK), opredeljena z vrednostjo pražnega cikla, za posamezne mikroorganizme v vzorcih bolnikov z drisko in vzorcih kontrolnih oseb razlikuje. V nalogi smo skušali opredeliti tudi uporabnost vzorca, ki smo ga pridobili s krtačastim brisom rektuma (krtačasti bris v transportnem gojišču Cary-Blair), saj zelo poenostavi pridobitev kužnin.&#8195; Hipoteze H1: Z molekularnim diagnosticiranjem bomo tudi pri kontrolni skupini oseb dokazali mikroorganizme, ki so potencialni povzročitelji infekcijske driske. Pričakujemo, da bomo pri kontrolnih osebah v vzorcih blata dokazali manjšo količino nukleinske kisline posameznih mikroorganizmov (opredeljeno z vrednostjo pražnega cikla) kot v vzorcih blata bolnikov z drisko ter opredelili tiste vrednosti pražnega cikla, pri katerih lahko sklepamo na klinično pomemben rezultat. H2: Število etiološko pojasnjenih drisk se statistično pomembno ne razlkuje, če za opredelitev povzročiteljev infekcijske driske (kot nadomestilo za vzorec blata) uporabimo krtačasti bris rektuma v transportnem gojišču Cary-Blair. Preiskovanci in metode V prospektivno raziskavo primerov in kontrol smo vključili 334 bolnikov obeh spolov, starih 18 let ali več in brez znanih imunskih okvar, ki smo jih od junija 2016 do avgusta 2017 zaradi driske pregledali na Kliniki za infekcijske bolezni in vročinska stanja (KIBVS) v Ljubljani in so zaradi resnosti bolezni potrebovali bolnišnično obravnavo (dnevna bolnišnica, bolnišnično zdravljenje), ter 154 kontrolnih oseb. Kontrolna skupina je vključevala osebe, ki v zadnjih štirih tednih pred vključitvijo niso imele driske in smo jih na KIBVS pregledali zaradi drugih težav, ter zdrave prostovoljce. Drisk, pridobljenih v bolnišnici, v naši raziskavi nismo spremljali. Bolnikov z drisko, ki so HIV-pozitivni ali s presajenimi organi zaradi večje homogenosti skupine v raziskavo nismo vključili. Za namen raziskave so bolniki oddali vzorec blata in bris rektuma, kontrolne osebe pa samo vzorec blata. Dodatne podatke smo pridobili na osnovi odgovorov v vprašalniku. Poleg mikrobioloških preiskav blata za dokaz povzročiteljev infekcijskih drisk – ki jih danes izvajamo rutinsko ter vključujejo kombinacijo metod kultivacije in neposrednega dokazovanja povzročiteljev z dokazom DNK ali antigenov – smo mikroorganizme dokazovali tudi z molekularnimi diagnostičnimi metodami. Pristopili smo »sindromsko«, saj smo s preiskavami hkrati dokazovali viruse, bakterije in parazite. Primerjali smo rezultate preiskav v vzorcih blata bolnikov in vzorcih blata kontrolnih oseb ter rezultate mikrobioloških preiskav vzorcev blata in brisov rektuma pri bolnikih. &#8195; Rezultati V končno analizo smo vključili 304 bolnike, pri katerih smo pridobili parne vzorce brisov rektuma in vzorcev blata, ter 154 oseb brez prebavnih težav (kontrolna skupina). Z uporabo molekularnega diagnosticiranja pri dokazovanju povzročiteljev infekcijske driske v blatu se je delež etiološko opredeljenih drisk z 51,3 % (sedaj uporabljene metode) povečal na 64,5 % (PCR). Z molekularnim testiranjem smo vsaj en patogen dokazali pri 196 od 304 bolnikov (stopnja pozitivnosti 64,5 %; 95 % IZ 58,8–69,9) in pri 20 od 154 kontrolnih oseb (stopnja pozitivnosti 13 %; 95 % IZ 8,1–19,3). V kontrolni skupini smo dokazali statistično pomembno manj patogenov (p < 0,005). Razmerje obetov (RO) je bilo 12,16 (95 % IZ 7,19–20,56). Več kot en patogen smo dokazali pri 23 od 304 bolnikov (7,6 %) (pri 21 bolnikih dva patogena, pri dveh bolnikih tri patogene) in pri dveh od 154 kontrolnih oseb (1,3 %). S primerjavo vrednosti pražnega cikla (Ct) pri bolnikih in pri kontrolnih osebah smo pri večini patogenov ugotovili višje vrednosti v kontrolni skupini (kar posredno odraža nižjo količino patogenov). Med bolniki in kontrolnimi osebami je bilo prisotno znatno prekrivanje v vrednostih Ct. Jasne mejne vrednosti Ct, ki bi bila tudi klinično pomembna, za posamezen patogen nismo uspeli opredeliti. Z uporabo molekularnega diagnosticiranja (hkratni PCR v realnem času) smo med 304 parnimi vzorci vsaj en patogen (v brisu rektuma in/ali v vzorcu blata) dokazali pri 226 bolnikih (74,3 %). Najmanj en patogen smo dokazali v 196 vzorcih blata (64,5 %) in v 193 brisih rektuma (63,5 %). Skladen rezultat je bil prisoten pri 241 od 304 bolnikov (79,3 %). Oba vzorca sta bila pozitivna za vsaj en patogen pri 163 bolnikih (53,6 %) in oba negativna pri 78 bolnikih (25,7 %). Pri 33 bolnikih (10,9 %) je bil pozitiven samo vzorec blata, pri 30 bolnikih (9,9 %) pa samo bris rektuma. Ob uporabi sindromskega molekularnega pristopa je bila celokupna občutljivost rektalnega brisa glede na vzorec blata kot zlati standard 83,2 % (95 % IZ 77,2–88,1), celokupna specifičnost pa 72,2 % (95 % IZ 62,8–80,4). Povprečne vrednosti Ct so bile za večino patogenov pri rektalnem brisu višje kot v vzorcu blata. Stopnja skladnosti rezultatov med kulturo in PCR (izražena s Cohenovo vrednostjo kappa (&#954;)) je bila dobra pri vzorcih blata (&#954; 0,798; 95 % IZ 0,730–0,867) in zmerna pri rektalnih brisih (&#954; 0,592; 95 % IZ 0,505–0,680). Ob uporabi molekularnega testiranja je bila občutljivost rektalnih brisov za bakterije 86,5 %, ob uporabi kulture za bakterije pa 61,4 %. Zaključki Z vpeljevanjem občutljivih molekularnih diagnostičnih metod je dilema glede pomena dokazane virusne nukleinske kisline v vzorcu blata – torej vprašanje, ali je dokaz potencialnega patogena v blatu dovolj za opredelitev povzročitelja driske – postala bistveno večja. V raziskavi smo ugotovili, da je razširjenost (prevalenca) enteropatogenov v blatu brezsimptomnih oseb pri odraslih manjša kot pri otrocih (glede na podatke iz literature). Podatki o količini DNK posameznih mikroorganizmov, ocenjeni z vrednostjo pražnega cikla (Ct), v vzorcih blata bolnikov z drisko in v vzorcih blata kontrolnih oseb ne pomagajo pri kliničnem tolmačenju izvidov molekularnih preiskav blata pri odraslih. V raziskavi smo pokazali, da je krtačasti bris rektuma v transportnem gojišču Cary-Blair sicer sprejemljiva alternativna možnost odvzema kužnine, če bolnik ob pregledu vzorca blata ne more oddati, a zahteva uporabo molekularnega diagnosticiranja. Ker so bili pri uporabi kulture bakterij brisi rektuma pomembno manj občutljivi od vzorcev blata, menimo, da pri brisih rektuma uporaba bakterijske kulture ni primerna.

• Language: Slovenian
• Keywords: infekcijska driska, molekularna diagnostika, Ct vrednosti, odrasli, bris rektuma
• Work type: Doctoral dissertation
• Organization: MF - Faculty of Medicine
• Year: 2019
• PID: 20.500.12556/RUL-107356
• COBISS.SI-ID: 299586816
• Publication date in RUL: 03.04.2019

Secondary language

• Language: English
• Title: USEFULNESS OF MOLECULAR METHODS FOR DETECTING PATHOGENS CAUSING INFECTIOUS DIARRHOEA IN ADULTS

Abstract

Background and aims With advanced molecular detection of enteropathogens and syndromic approach, significantly more diarrhoea causing pathogens are detected, but there is also concern of overdiagnosis. Interpretation of microbiological findings can be challenging and with two or more detected pathogens in the stool together with limited data on prevalence of enteropathogens in the stool of asymptomatic healthy individuals, the decision what to treat, when appropriate, can be difficult. Very few data exist on the prevalence of enteropathogens in healthy asymptomatic adult population, either using conventional methods or molecular testing and only few case-control studies in adults have been published. Data from case-control studies are scarce in the literature. Little is known about Ct values (cycle of threshold, Ct) of individual pathogen in stool samples from adult patients with diarrhoea and healthy asymptomatic individuals and the role of Ct value in interpretation of microbiological test results. A stool sample is a sample of choice for microbiological testing of enteric pathogens causing diarrhoea. However, in certain circumstances stool is not readily available and a rectal swab can be a more practical alternative. Most of the studies were done in children and very few data exist for the adult population with regard to rectal swab use for diagnosing pathogens causing infectious diarrhoea. The aim of the prospective case control study was to evaluate which pathogens can be detected in the stool of adult patients and asymptomatic adult individuals and whether Ct values - as indirect measure of pathogen load - differ between those two groups. A prospective observational study was performed to evaluate the diagnostic performance of flocked rectal swab specimens using the syndromic molecular approach to determine the aetiology of diarrhoea in adults. &#8195; Hypotheses H1: With molecular methods, enteropathogens which can potentially cause infectious diarrhoea will be found in the stool samples of asymptomatic individuals. We expect that in the stool of individuals in control group the relative load of certain pathogen (expressed with Ct value) will be lower (higher Ct value) compared to the patient group. We further anticipate that the clinically meaningful »cut-off« Ct values for certain pathogen will be determined. H2: The use of of flocked rectal swab specimens (in transport media Cary-Blair) as alternative to stool samples in determination of the aetiology of infectious diarrhoea in adults does not significantly reduce the number of etiologically confirmed diarrhoea cases. Patients and methods The prospective case-control study was conducted at the University Medical Centre Ljubljana, Department of Infectious Diseases, using syndromic detection methods for viruses, bacteria and parasites. 334 adult patients (age 18 years or older) with diarrhoea (defined as &#61619; 3 loose stools in 24 hours) who presented at the emergency room, were enrolled in the study period between June 2016 and August 2017. Patients with known immunodeficiencies (HIV, transplant patients) and patients with hospital acquired diarrhoea were not included in the study. Healthy volunteers and patients presented at the clinic due to non-diarrhoeal related illnesses were recruited in the control group. Individuals from the control group who were enrolled in the study came from the same Slovenian region as patients and were enrolled in the study within the same time frame. Individuals, having diarrhoea within 4 weeks before the enrolment in the study were not eligible to participate. All of the included patients and individuals in control group filled out the questionnaire. Additional epidemiological, clinical and laboratory data were obtained by a questionnaire and with a review of medical charts. Informed consent was obtained from all participants included in the study. Stool samples and rectal swabs were collected from adult patients with diarrhoea, and stool samples only from adult individuals without gastrointestinal problems (control group). The stool samples were analysed by multiplex real-time PCR for selected target organisms (7 bacteria, 4 viruses, 2 protozoa) and conventional microbiological testing (culture, antigen detection, microscopy). Positivity rates in case group and in control group and OR were calculated and Ct values were measured. We compared the performance of rectal swabs to stool samples as the gold standard in determining viral, bacterial and protozoal pathogens using real-time multiplex PCR as well as standard stool culture. Results Paired samples of stool and rectal swab specimens were collected from 304 adult patients with diarrhoea. In control group there were 154 individuals without gastrointestinal symptoms. Case-control analysis was performed in a total of 304 stool samples from patients and 154 samples from control group. With molecular testing positivity rate of 64.5 % (95%CI 58.8-69.9) for any pathogen was observed in patients and 13 % (95% CI 8.1-19.3) in control group (p<0.005). Odds-ratio (OR) for all pathogens was 12.16 (95% CI 7.19-20.56). More than one pathogen was detected in 23 out of 304 patients (7,6 %) (in 21 patients two pathogens were detected and in 2 patients three pathogens were detected simultaneously), and in 2 out of 154 individuals in control group (1,3 %). The mean Ct values for individual pathogen were higher in control group, but with overlap and no clinically signficant cut-off value could be established. Rectal swab was evaluated on paired samples of stool and rectal swab specimens from 304 adult patients with diarrhoea. At least one pathogen was detected (in rectal swabs and/or stool samples by PCR) in 226 patients (74.3%). Among all 304 paired samples, 196 (64.5%) of stool samples and 193 (63.5%) of rectal swabs were positive for at least one pathogen when molecular diagnostic tests were applied. The concordant results were present in 241 (79.3%) patients: both rectal swab and stool were positive for any pathogen in 163 (53.6%) patients. In 33 (10.9%) patients only stool was positive and in 30 (9.9%) patients only rectal swab was positive. Overall sensitivity of rectal swab samples in syndromic molecular approach was 83.2% (95% CI 77.2-88.1) and specificity 72.2% (95% CI 62.8-80.4). Mean threshold cycle (Ct) values for most pathogens were higher in rectal swab specimens than in stool specimens. Agreement between culture and PCR was good in stool samples (kappa 0.798, 95% CI 0.730-0.867) but moderate for rectal swabs (kappa 0.592, 95% CI 0.505-0.680) For bacteria sensitivity was 86.5% (95%CI 79.5-91.8) when PCR was performed and 61.4% (95% CI 52.4-69.9) when culture for bacteria was performed. Conclusions Our study showed that the positivity rate in asymptomatic individuals is not as high as anticipated based on other studies which included all age categories and not just adults. Our data show that the rate of detected pathogen in the stool of asymptomatic individuals is lower in adult population, therefore the clinical significance of detected pathogen by molecular testing is probably of more value in adults than in children. According to results of our study, the Ct values are not helpful in clinical interpretation of microbiological result in adult population. Our results indicate that rectal swabs can be used in diagnostics of diarrhoea in adults when stool specimens are not available or/and rapid aetiological determination is needed. However, rectal swabs should be analysed using molecular approach. The mean Ct value for most pathogens is higher in rectal swab specimens than in stool specimens.

• Keywords: infectious diarrhea, molecular diagnostics, Ct values, adults, rectal swab