| Abstract | Svi kirurški zahvati, pa tako i kardiokirurški zahvati, praćeni su određenim postotkom mortaliteta operiranih osoba. Postotak smrtnosti kod postavljanja srčanih premosnica varira ovisno o državi i medicinskom centru, a kreće se u rasponu od 2,0-4,6%. U zadnja dva desetljeća razvijen je veći broj bodovnih sustava koji procjenjuju i kvantificiraju čimbenike rizika, to jest predviđaju mortalit kod kardiokirurških zahvata. Europski sustav za procjenu rizika kardiokirurških zahvata (EuroSCORE) razvijen je za predikciju mortaliteta kod postavljanja srčanih premosnica i zamjena srčanih zalistaka. Međutim, premda je uporabna vrijednost Euro-SCORE-a neupitna, postoji još dosta prostora za njegovo unapređenje. Analiza britanske baze podata s 16.619 bolesnika s obavljenim postavljanjem srčanih premosnica pokazala je da je učinkovitost Euro-SCORE-a u predikciji mortaliteta, mjerena metodom površine ispod ROC krivulje (AUC) svega 0,743, s 95% granicama pouzdanosti u rasponu od 0,721 do 0,765. Predložen je čitav niz novih varijabli kao nadopuna Euro-SCORE sustavu, a uglavnom se koriste različite biokemijski parametri. U našoj studiji, provedenoj na 136 bolesnika kojima je izvršeno postavljanje srčanih premosnica, uspoređivana je prediktivna snaga adističkog i logističkog oblika EuroSCORE bodovnog sustava s prediktivnim vrijednostima serumski određenog NT-proBNP. Ukupno je bilo 5 umrlih ispitanika (3,68%). Uporabom rpart programa utvrđeno je kako preoperativne serumske vrijednosti NT-proBNP, u kombinaciji s dobi ispitanika, daju bolju procjenu relativnog rizika umiranja u odnosu na aditivni ili logistički model EuroSCORE bodovnog sustava. Superiornost kombinacije preoperativnih vrijednosti serumski određenog NT-proBNP i dobi bolesnika vjerojatno potječe od toga što ti parametri sadrže dodatnu informaciju koja nedostaje EuroSCORE-u. Rezultati upućuju kako NT-proBNP daleko bolje opisuje stanje kardiovaskularnog sustava od EuroSCORE-a. Dob, koja je korelirana s određenim faktrorima rizika, poput šećerne bolesti, dodatno nadopunjava prediktivnu snagu NT-proBNP. Nadalje, konstruiran je vrlo jednostavan algoritam za razvrstavanje bolesnika u rizične podskupine: 1. Ako se relativni rizik umiranja tijekom operacije i u periodu od 30 dana za SVE ispitanike označi s brojem 1, tada je najvažnija varijabla koja razdvaja ispitanike u dvije podskupine s najvećom razlikom u relativnom riziku umiranja serumski određena preoperativna vrijednost NT-proBNP. Kod ispitanika s vrijednošću NT-proBNP-a manjom od 1433,5 relativni rizik umiranja može se označiti s brojem 0,54, dok se kod ispitanika s vrijednostima NT-proBNP >= 1433,5 relativni rizik umiranja može označiti s brojem 2,9. 2. Kod podskupine ispitanika s vrijednošću NT-proBNP-a manjom od 1433,5 relativni rizik umiranja se može još preciznije odrediti uporabom varijable dob. Granična vrijednost varijable dob je 68,5 godina. Kod ispitanika s dobi < 68,5 godina relativni rizik se može označiti s brojkom 0,22, dok je kod ispitanika u dobi >= 68,5 relativni rizik 1,43. Dobiveni algoritam upućuje kako je kombinacija preoperativnih vrijednosti serumski određenog NT-proBNP i dobi bolesnika vrlo jednostavan, jeftin i pouzdan način procjene relativnog rizika umiranja kod bolesnika s kirurškom revaskularizacijom, to jest postavljanjem srčanih premosnica. |
| Abstract (english) | All surgical procedures including cardiac surgery are followed by some rate of mortality among operated patients. Mortality rate of patients with aortocoronar bypass varies between hospitals, depending on state where procedure took place and of clinics in that very state, and the range is between 2.0 and 4.6%. In last two decades are followed by developing of some numbers of score systems wich evaluate and quantify risk factors, in other way, thay predict mortality among cardiasurgical procedures. European system for cardiac opertive risk evaluation (EuroSCORE) has been developed for prediction of mortality during bypass procedure and replacements of valvulas. Althoug the usage value of EuroSCORE is unquestioned, there is still lot of space for its improvement. The analyse of british data base with 16 619 patients withg bypass procedure has shown that efficiency of EuroSCORE in mortality prediction, measured by the area method under the ROC curve (AUC) is only 0.743, with 95% reliability limits in range of 0.721 to 0.765. The whole range of new variables have been recommended as supplement to EuroSCORE, but usually various biochemical parameters are used. In our study conducted on 136 patients with bypass procedure predictive power of aditive and logistic form of EuroSCORE scoring system has been compared with predictive values of serume determined NT-proBNP. It has been totally 5 deacesed examed patients (3.68%). Using the rpart programme have been estimated that preoperative serume values NT-proBNP, in combination with the age of examed patients are giving better evaluation of relative risk of deacesing compared to aditive or logistic model of EuroSCORE system. The main strength of preoperative determined serume values of NT-proBNP and pateint age most probably adding significant information that are not currently used in EuroSCORE scoring system. Preliminary results showed with NT-proBNP are stronger correlate with current cardiovascular status then EuroSCORE scoring system. Age that significantly correlates with some risk factors, such as diabetes, additionaly support predictive value of serume determined NT-proBNP. Furthermore, clear and simple algorhythm has been formed for successfull pateint selection in risk groups: 1. If motalitiy relative risk during the surgery and 30 days after the procedure is predetermined as 1 then the serum value of preoperative NT-proBNP is the most important variable that divide participants in 2 subgroups with greatest difference in mortality relative risk. In subgruop of participants with serum value of NT-proBNP <1433.5, determined mortality relative risk is 0.54, and in subgroup of participants with serum value of NT-proBNP >=1433.5, determined mortality relative risk is 2.9. 2. In pateint subgroup with NT-proBNP values lower than 1433.5, death mortality relative risk can be more precisely determined using age as additinal scoring factor. Borderline value of age scoring factor is 68.5 years. In participants with age lower then 68.5 years, relative risk is determined as 0.22, and in those >= 68.5 years is 1.43, respectively. This algorhythm shows that combination of preoparative serume values of NT-proBNP and patient age is simple, inexpensive and reliable mortality risk estimation factor in patients after CAGB. |
