The role of clinical pharmacist in the management of clinically relevant drug-drug interactions in patients with chronic heart failure

Interakcije med zdravili predstavljajo posebno kategorijo neželenih učinkov, pri katerih učinki enega zdravila vplivajo na učinke drugega zdravila. Te se lahko izrazijo kot zmanjšana učinkovitost ali toksičnost, ki lahko poveča obolevnost ali smrtnost pri bolnikih. Interakcije med zdravili predstavljajo vzrok 1% bolnišničnih sprejemov in 16% sprejemov zaradi neželenih učinkov zdravil. Staranje populacije vodi v povečano število sočasnih obolenj in posledično povečano kompleksnost zdravljenja. Implementacija navodil iz smernic za obravnavo bolezni priporoča sočasno zdravljenje s številnimi zdravili. To poveča verjetnost za nastanek interakcij med zdravili, saj ta narašča s številom sočasno predpisanih zdravil. Pogosto uvajanje novih zdravil in odobritve novih indikacij za zdravila v klinični praksi otežijo predpisovalcem prepoznavanje interakcij med zdravili. V pomoč so dostopni različni sistemi, tako v elektronski kot v knjižni obliki, ki služijo prepoznavanju neželenih reakcij in olajšajo odločitve pri predpisovanju različnih kombinacij zdravil. Večina teh sistemov ima ključno pomanjkljivost, saj zaznavajo vse potencialne interakcije med zdravili, tudi tiste, ki ne vodijo v klinično pomembne izide. Poleg tega ovrednotijo interakcije zgolj med posameznimi pari zdravil ne upoštevajoč ostalo terapijo. Hkrati ne upoštevajo sočasnih obolenj, laboratorijskih parametrov ter značilnosti bolnika, ki so ključni za klinično pomembnost interakcij. Zato predstavlja klinična ocena pomemben dejavnik pri oceni razširjenosti interakcij med zdravili, ki lahko vodijo v neželene učinke. Srčno popuščanje je med največjimi izzivi sodobne medicine zaradi svoje pogostosti, sočasnih obolenj in visoke obolevnosti ter umrljivosti. Njihova obravnava je zaradi komorbidnosti, starostno pogojenih sprememb v funkciji organov in potrebe po nadzoru zelo kompleksna in povezana s številnimi izzivi. Zdravljenje z zaviralci renin-angiotenzin-aldosteronskega sistema, ki naj jih po priporočilih prejemajo vsi bolniki s sistolično disfunkcijo levega prekata in so primerni za zdravljenje komorbidnosti pri bolnikih z ohranjeno sistolično funkcijo levega prekata, lahko ob kronični ledvični bolezni povzroči akutno poslabšanje ledvične funkcije ali hiperkaliemijo. Individualno prilagajanje farmakološkega zdravljenja je tako zahtevno in vključuje skrbno spremljanje bolnika, njegovih kliničnih parametrov ter laboratorijskih izvodov. Klinični farmacevti igrajo pomembno vlogo pri spremljanju zdravljenja, svetovanju zdravnikom in bolnikom ter pripomorejo k zmanjšanju števila zapletov in klinično pomembnih izidov. Cilj doktorske naloge je bil ovrednotiti vpliv svetovanja kliničnega farmacevta glede klinično pomembnih interakcij med zdravili pri bolnikih s srčnim popuščanjem ob sprejemu, na zmanjšanje njihove razširjenosti ob odpustu ter vpliv na dolgoročne izide. V prvem delu smo izvedli sistematični pregled literature. Pregledali smo tri podatkovne baze z namenom identificirati publikacije, ki so ovrednotile uporabo elektronskih sistemov za zaznavanje interakcij med zdravili pri različnih skupinah bolnikov. Kljub velikemu številu in široki uporabi elektronskih sistemov je bilo publikacij, ki so ustrezale našim kriterijem relativno malo. Ti sistemi zaznavajo zgolj potencialne reakcije med zdravili. Njihovo poročanje temelji na literaturnih podatkih in definiranih algoritmih, kar vodi v generiranje velikega števila poročil, ki pogosto nimajo kliničnega pomena. To lahko povzroči, da predpisovalci spregledajo interakcije, ki lahko povzročijo neželene reakcije pri bolnikih. Kljub temu služijo kot uporabno orodje, saj jih proizvajalci redno posodabljajo glede na nove literaturne podatke. Frekvenca teh posodobitev in tudi kvaliteta samih poročil se med različnimi sistemi razlikuje. V okviru sistematičnega pregleda smo publikacije razdelili v tri podskupine. Prva opisuje prevalenco potencialnih interakcij določenih z enim elektronskim sistemom, druga primerja dva ali več sistemov med sabo in tretja vključuje poleg identifikacije potencialnih interakcij med zdravili z elektronskim sistemom tudi klinično oceno strokovnjakov. Publikacij iz zadnje skupine je najmanj, za vse pa je značilna velika razlika v številu določenih interakcij z elektronskim sistemom in oceno strokovnjakov. Pregledni članek smo objavili v reviji European Journal of Clinical Pharmacology. Svoje raziskovanje smo nadaljevali z retrospektivno raziskavo, s katero smo s pomočjo elektronskega sistema želeli ovrednotiti število potencialnih interakcij med zdravili ob sprejemu in ob odpustu pri bolnikih s srčnim popuščanjem in kronično obstruktivno boleznijo pljuč. Za oceno prevalence smo uporabili elektronski sistem Lexi-Interact®, ki klasificira interakcije v skupine C, D in X glede na ukrepe, ki jih je potrebno sprejeti ob njihovi prisotnosti. Tega smo izbrali zaradi dostopnosti in ker se je na podlagi sistematičnega pregleda literature izkazal za enega izmed najbolj zanesljivih glede na literaturo, ki podpira oceno, kvaliteto poročil in priporočil ter na pogostost posodobitev. Raziskava je potekala na Univerzitetni kliniki Golnik. Vključili smo vse zaporedne bolnike, ki so bili odpuščeni ali so umrli v obdobju šestih mesecev z diagnozo srčnega popuščanja ali kronične obstruktivne bolezni pljuč. Iz bolnikovih odpustnih listov smo dobili podatke o zdravilih, ki so jih bolniki prejemali ob sprejemu in ob odpustu. S pomočjo elektronskega sistema Lexi-Interact smo ovrednotili celokupno število interakcij ob sprejemu in ob odpustu ter število interakcij različnega tipa. Mediana števila interakcij se ob sprejemu in ob odpustu ni razlikovala. Statistično signifikantno se je povečalo število interakcij tipa C in X (p=0.01) ob odpustu, glede na njihovo število ob sprejemu. Najbolj pogost tip interakcij je bil tip C in sicer v 373 od 485 primerov ob odpustu. Za ta tip interakcij se priporoča zgolj spremljanje terapije. Sledila sta tipa D v 98 in X v 45 primerih. Število interakcij pri bolnikih je bilo značilno povezano s številom diagnoz in številom zdravil, ki so jih prejemali bolniki. Ob uporabi elektronskega sistema smo ugotovili, da ta generira veliko število poročil o interakcijah, ki niso klinično relevantne, saj pri njihovi identifikaciji ne upošteva značilnosti bolnika. Določene interakcije, ki bi se lahko izrazile s pomembnimi kliničnimi izidi ovrednoti kot tip C, medtem ko druge rangira kot tip X, kontraindikacije, čeprav gre pri veliki večini bolnikov za dobro prenosljive kombinacije zdravil, ki so del rutinske klinične prakse. Rezultati retrospektive raziskave so omogočili oceno velikosti vzorca za prospektivno raziskavo. Članek iz tega dela rezultatov smo objavili v reviji Archives of medical Science. Na podlagi ugotovitev iz retrospektivnega dela raziskave smo se v naslednjem delu osredotočili na klinično relevantne interakcije pri bolnikih s srčnim popuščanjem. V dvojno slepi randomizirani raziskavi smo ovrednotili vpliv intervencije kliničnega farmacevta na število klinično relevantnih interakcij ob odpustu. Raziskava je bila registrirana v bazi www.clinicaltrials.gov. V tri-mesečnem obdobju smo pregledali vse bolnike, ki so bili sprejeti na Univerzitetni kliniki Golnik z diagnozo srčnega popuščanja. Vključili smo bolnike, ki so imeli ob sprejemu vsaj eno klinično relevantno interakcijo med zdravili. Oceno klinične pomembnosti je za vsakega bolnika posebej podala ekspertna skupina zdravnikov, na podlagi ocene elektronskega sistema, prisotnosti sočasnih obolenj in laboratorijskih podatkov bolnika. Interakcije med zdravili so bile ovrednotene kot klinično relevantne, če so lahko vodile v klinično pomembne izide, kot je na primer poslabšanje bolezenskega stanja ali hospitalizacija. Osebnih podatkov, razen starosti in spola, skupina ekspertov ni prejela. Ekspertna skupina je farmacevtu sporočila prisotnost klinično relevantne interakcije med zdravili, na podlagi katere je ta pripravil opozorilo in nasvet lečečemu zdravniku. Podatke je nato posredoval drugemu farmacevtu na Univerzitetni kliniki Golnik, ki je na podlagi randomizacijskega ključa razvrstil bolnike v kontrolno in interventno skupino. Lečeči zdravniki so za bolnike, ki so bili razvrščeni v kontrolno skupino, prejeli zgolj generični nasvet, za bolnike v interventni skupini pa specifično priporočilo glede prisotne klinično relevantne interakcije. Število interakcij je bilo ocenjeno ponovno ob odpustu. V interventni skupini bolnikov se je celokupno število klinično pomembnih interakcij od sprejema do dopusta statistično signifikantno zmanjšalo. Ob odpustu je bilo v interventni skupini manjše število celokupnih interakcij kot v kontrolni skupini. Vpliva intervencije na dolgoročne izide smrtnosti in ponovnih hospitalizacij nismo zaznali. Članek iz tega dela raziskave je bil objavljen v reviji International Journal of Cardiology, del razultatov te raziskave pa je bil vključen v publikacijo, ki je bila poslana v recenzijo v European Journal of Internal Medicine. Doktorska disertacija predstavlja pomemben doprinos k metodologiji in oceni klinično pomembnih interakcij med zdravili pri bolnikih s srčnim popuščanjem. Naši rezultati tudi kažejo na pomembnost vloge kliničnega farmacevta v tem procesu. Večina raziskav na tem področju se je usmerila na poročanje števila potencialnih interakcij z različnimi elektronskimi sistemi. Le majhno število raziskav je vključilo tudi oceno klinika in s tem ovrednotilo interakcije med zdravili, ki lahko vodijo v klinično pomembne izide. To smo pokazali z našim pregledom literature. Večina publikacij predstavlja to kot omejitev, saj je število potencialnih interakcij na posameznega bolnika v večini primerov zelo visoko. To smo opazili tudi mi z retrospektivnim delom raziskave, kjer smo zaznali kar 5604 potencialnih interakcij ob odpustu na skupini 778 bolnikov. Zato smo se v prospektivnem delu raziskave osredotočili zgolj na klinično relevantne interakcije ter na vlogo kliničnega farmacevta pri njihovi obravnavi. Kot je bilo pričakovano, je število klinično relevantnih interakcij na bolnika značilno manjše kot število potencialnih interakcij pri primerljivi skupini bolnikov. Z raziskavo smo pokazali pozitiven vpliv svetovanja kliničnega farmacevta na zmanjšanje števila klinično pomembnih interakcij ob odpustu. Te izide smo dosegli s skoraj avtomatiziranim sistemom poročanja, brez osebnega stika farmacevta z zdravnikom. Učinki bi bili verjetno še večji v redni klinični praksi zaradi osebnih stikov tako z zdravniki kot tudi z bolniki. Po drugi strani pa to zahteva več časa in posledično obravnavo manjšega števila bolnikov. Vpliva na dolgoročne klinične izide zaradi majhnega vzorca bolnikov nismo uspeli zaznati. Zato bi bile potrebne nadaljnje raziskave na večjem številu bolnikov, ki bi lahko pokazale pozitiven vpliv na zmanjšanje števila hospitalizacij in smrtnosti, saj to predstavlja prihranek za zdravstveni sistem.Drug-drug interactions represent a special category of adverse drug reactions in which the effects of one drug influence the effects of the other. These can be expressed as reduced effectiveness or toxicity which can lead to increased morbidity or mortality of patients. Drug-drug interactions represent a cause to 1% of hospital admission and 16% of admissions due to adverse drug reactions. Ageing of population leads to comorbidity and therefore increases the complexity of treatment. Guideline implementation requires treatment with several concomitant medications which increases the occurrence of drug-drug interactions as their probability increases with the number of prescribed drugs. Frequent launches of new drugs and approval of new indications for marketed medicines make recognition of occurrence of drug-drug interactions more difficult for health care professionals. To cope with that, several drug-drug interactions screening programs and databases have been developed and implemented as clinical decision support systems. Most of these systems have limitation that they report drug-drug interactions which do not lead to clinically relevant outcomes and the drug-drug interactions between specific drug pairs regardless of the concomitantly prescribed drugs. When reporting, these systems do not consider concomitant diseases, laboratory data and patients characteristics. Therefore, clinicians’ assessment plays the important role in the evaluation of the occurrence of drug-drug interactions which may cause adverse drug events. Patients with chronic heart failure are characterized with high morbidity and several concomitant diseases. Therefore is the treatment frequently complex and challenging specifically due to age related organ changes. Based on the guideline recommendations these patients are co-prescribed several medications which can lead to clinically relevant outcomes due presence of concomitant diseases. Such medications are inhibitors of renin-angiotensin-aldosterone system which are according to the guideline recommendation prescribed to all patients with left ventricular systolic dysfunction and can in presence of chronic kidney disease lead to worsening of renal function and hyperkalemia. Monitoring of the laboratory parameters, clinical outcomes and dose adjustments are required in these cases. Clinical pharmacists play an important role in monitoring of the treatment, counseling to physicians and patients and contribute to reduced number of complications and clinically relevant outcomes. The aim of this thesis was to evaluate the impact of counseling of clinical pharmacist about the presence of clinically relevant drug-drug interactions in patients with heart failure on admission to the hospital, on their reduction at discharge and the long term outcomes. In the first part we preformed systematic review of the available publications. We screened three databases aiming to identify the publications, which have evaluated the use of electronic systems available for identification of drug-drug interactions in a broad spectrum of patients. Despite the large number and broad usage of these systems only a small number of publications fulfilled our selection criteria. These systems identify only potential drug-drug interactions. The reports are based on the literature data and pre-defined algorithms which lead to generation of the large number of reports, which are often of low clinical importance. This causes clinicians to overlook the drug-drug interactions which could lead to the adverse events. Despite that, they represent a useful tool as their developers regularly update the information with newly available data from the literature. Frequency of updates as well as the quality of the reports differs significantly among the systems. In the scope of our systematic review we have classified the publications into three groups. The first group of publications focuses on the prevalence of potential drug-drug interactions evaluated with one electronic system, the second one compares two or more systems between each other and the third includes also clinician’s assessment in addition to the assessment with the electronic system. The smallest number of publications was identified in the third group. All of them report high difference in number of identified drug-drug interaction between the electronic system and clinician’s assessment. The systematic review article was published in European Journal of Clinical Pharmacology. We continued our research with retrospective study in which we used the electronic system to evaluate the number of potential drug-drug interactions in patients with heart failure and chronic obstructive pulmonary disease on hospital admission and at discharge. For evaluation of the prevalence we used Lexi-Interact system, which classifies potential drug-drug interactions as types C, D and X according to the measures which have to be taking in the case of their occurrence. This system was chosen due to its availability and as it was as per systematic review we performed ranked high regarding the reliability of literature which supports the reports as well as their quality and the frequency of the updates. The research was conducted in the University Clinic in Golnik. Patients, who were consecutively discharged or died in the period of 6 months with diagnosis of heart failure and chronic obstructive pulmonary disease, were included in the study. The information about the prescribed medications on admission and at discharge was obtained from patients’ discharge letters. The total number and the type of potential drug-drug interactions were assessed with Lexi-Interact system. There was no significant difference between the median number of potential drug-drug interactions on admission and at discharge. There were statistically significantly more potential drug-drug interactions type C and X (p=0.01) at discharge than on admission. The most common type of potential drug-drug interactions at discharge was type C in 373 out of 485. These require only monitoring of the therapy. Types D and X were observed in 98 and 45 cases, correspondingly. The number of potential drug-drug interactions in patients strongly correlated with the number of diagnoses and the number of medicines prescribed to patients. The system reports high number of potential drug-drug interactions which are not clinically relevant as it does not consider patient’s characteristics. Some interactions classified as type C could lead to clinically relevant outcomes, while on the other side contraindication or type X is reported for drug combination which is part of daily clinical practice. The outcomes of the retrospective study were used as basis for the sample size estimation of the prospective study. The results of the retrospective research were published in Archives of Medical Science. Based on the findings of the retrospective research we focused our next research on clinically relevant drug-drug interactions in patients with heart failure. We designed double blind randomized controlled trial in which we evaluated the impact of pharmacist’s intervention on the number of clinically relevant drug-drug interactions at patients’ discharge. The study was registered in www.clinicaltrials.gov. Patients with diagnosis of heart failure who were admitted to the University Clinic Golnik in the three months period were screened for the inclusion. Patients with at least one clinically relevant drug-drug interaction were included in the study. The assessment of clinical relevance was performed for each patient individually by the panel of experts based on the assessment of the electronic system, comorbidities and patients’ laboratory data. Drug-drug interactions were evaluated as clinically relevant if their occurrence could lead to clinically relevant outcomes such as worsening of medical condition or hospitalization. The panel was blinded for personal information of the patients except for the age and gender. In case of presence of clinically relevant drug-drug interaction the panel of experts submitted a report to clinical pharmacist who prepared a note and recommendation to the treating physicians. These data were sent to the the other pharmacist in the University Clinic Golnik who classified the patients into control and intervention group based on the randomization sequence. Treating physicians whose patients were randomized into control group received only generic advice, while for patients in intervention group physicians received also the recommendation about the present drug-drug interaction. The number of clinically relevant drug-drug interaction was evaluated again at discharge. A statistically reduction of total number of clinically relevant drug-drug interactions from admission to discharge was observed in intervention group. There was also significantly lower number of clinically relevant drug-drug interactions at discharge in intervention group as in control group. The impact on long term outcomes in the number of re-hospitalizations or death has not been observed. The article from findings of this research was published in the International Journal of Cardiology, part of the results was included in the publication submitted to European journal of Internal Medicine. The thesis represents an important contribution to the evaluation of the methodology of the evaluation of occurrence of clinically relevant drug-drug interactions in patients with heart failure. The outcomes emphasize the importance of the role of clinical pharmacist in this process. Most publications in this area focused on reporting of potential drug-drug interactions with electronic systems. Only a small number of researches included also the clinician’s assessment and evaluated the drug-drug interactions which may lead to clinically relevant outcomes. This was presented with systematic review. Most of these publications represent this as a major limitation, as the number of potential drug-drug interactions is very high. We have showed this in our retrospective study as we identified 5604 potential drug-drug interactions on the group of 778 patients. Due to this reason we have focused our further research on clinically relevant drug-drug interactions and the role of pharmacists in their management. As expected the occurrence of these is much lower in the comparable group of patients. The outcomes of this research have showed positive impact of pharmacist’s counseling on the reduction of clinically relevant drug-drug interactions at discharge. The results were achieved with almost authomated system, without personal contact between prescriber and the pharmacist. The impact could be even greater in the standard clinical practice due to contact with doctors and patients. This is more time consuming and therefore allows management of fewer number of patinets. Due to small samp