| Abstract | Cilj istraživanja: usporediti učinkovitost i štetnost pojedinih vrsta pripravaka enteralne prehrane, kao i usporediti enteralnu prehranu u odnosu na totalnu parenteralnu prehranu i pristup liječenja bez nutritivne potpore u bolesnika s akutnim pankreatitisom. Dodatni cilj uključuje i procjenu učinkovitosti i štetnosti enteralne prehrane u ovisnosti o težini bolesti. Materijali i metode: temeljem opsežne i sustavne pretrage literature, ovaj sustavni pregled obuhvaća sva randomizirana klinička istraživanja u kojima se određeni oblik enteralne prehrane uspoređuje s drugim oblikom nutritivnog pristupa liječenju (druga enteralna formula, totalna parenteralna prehrana ili pristup bez nutritivne potpore) akutnog pankreatitisa kao kontrolom. Iz studija koje ispunjavaju uključne kriterije prikupljeni su podaci o karakteristikama istraživanja te primarnim i sekundarnim ishodima. Primarni ishodi obuhvaćaju mortalitet bilo kojeg uzroka, sindrom sustavnog upalnog odgovora, organsko zatajenje i nuspojave, dok sekundarne ishode čine lokalne septične komplikacije, ostale lokalne komplikacije, ostale infektivne komplikacije, serumska koncentracija C-reaktivnog proteina i duljina trajanja hospitalizacije. Analiza je provedena računalnim programom Review Manager inačica 5.2, primjenom fixed-effect i random-effects modela meta-analize. Rezultati su izraženi kao relativni rizik (RR) za dihotomne podatke i razlika aritmetičkih sredina (MD) za kontinuirane podatke, u oba slučaja s 95% intervalima pouzdanosti. Analiza je provedena po intention-to-treat principu. Rezultati: pretraživanjem literature je ukupno pronađeno 1308 publikacija. Izostavljanjem dvostrukih rezultata i publikacija koje nisu randomizirana istraživanja ili nisu povezane s tematikom sustavnog pregleda, preostala je 41 studija, od kojih je temeljem uključnih kriterija 25 uključeno u sustavni pregled i meta-analizu. Imunonutricija u odnosu na kontrolu značajno smanjuje mortalitet bolesnika (RR 0, 49 ; 95% CI 0, 29 do 0, 80), dok značajna razlika u ostalim ishodima s obzirom na manjak prikaza podataka u studijama nije dokazana. Sličan je učinak utvrđen i za podgrupu bolesnika s teškim akutnim pankreatitisom. Analiza enteralne prehrane s probioticima ukazala je na značajno manju učestalost organskog zatajenja (RR 0, 77 ; 95% CI 0, 62 do 0, 97) i lokalnih septičnih komplikacija (RR 0, 60 ; 95% CI 0, 41 do 0, 89), ali isključivo primjenom fixed-effect modela zbog značajne heterogenosti među studijama. Pojavnost ostalih infektivnih komplikacija bila je značajno manja u skupini liječenoj probioticima (RR 0, 56 ; 95% CI 0, 32 do 0, 98). U jednoj studiji ove analize utvrđen je značajno veći broj ozbiljnih nuspojava, koje su rezultirale značajno većom učestalošću organskog zatajenja i smrtnosti. Zbog visokog rizika pristranosti i manjkavosti u dizajnu studije, provedena je post hoc analiza s isključenjem rezultata iste. Time je utvrđen dodatni blagotvorni učinak probiotika na spomenute ishode. Primjena semielementarnih formula značajno je smanjila smrtnost (RR 0, 33 ; 95% CI 0, 20 do 0, 54), učestalost organskog zatajenja (RR 0, 42 ; 95% CI 0, 32 do 0, 56), broj lokalnih septičnih komplikacija (RR 0, 40 ; 95% CI 0, 30 do 0, 55) i ostalih lokalnih komplikacija (RR 0, 48 ; 95% CI 0, 34 do 0, 67). Ostali su ishodi bili slični u obje grupe ispitanika. Procjenu učinkovitosti polimernih enteralnih formula bilo je moguće provesti na temelju samo dvije studije, bez utvrđenog značajnog učinka na sve analizirane ishode. Slični rezultati na većinu kliničkih ishoda bolesti dobiveni su analizom dvije studije koje su enteralnu prehranu obogaćenu vlaknima uspoređivale s kontrolom, uz statistički značajno kraće vrijeme hospitalizacije (MD -9, 28 ; 95% CI -13, 21 do -5, 35) i manji broj ostalih lokalnih komplikacija (11/28 naspram 24/32 ; P=0, 008). Samo je jedna studija istraživala imunonutriciju obogaćenu probioticima i vlaknima, zbog čega meta-analizu nije bilo moguće provesti, a opisuje značajno kraće vrijeme hospitalizacije (P=0, 01). Analizom podgrupa utvrđeno je da je primjena bilo kojeg oblika enteralne prehrane značajno učinkovitija od totalne parenteralne prehrane u smanjenju smrtnosti (RR 0, 33 ; 95% CI 0, 21 do 0, 54) i organskog zatajenja (RR 0, 39 ; 95% CI 0, 30 do 0, 52), uz podjednak broj nuspojava liječenja. Enteralna prehrana pokazala se boljom od pristupa bez nutritivne potpore u smanjenju mortaliteta (RR 0, 50 ; 95% CI 0, 29 do 0, 86), dok zbog manjka prikazanih podataka nije bila moguća meta-analiza preostalih primarnih ishoda. Sekvencijskom analizom studija za one analize ishoda mortaliteta koje su pokazale statistički značajan rezultat, utvrđena je dovoljna količina i snaga dokaza u slučaju usporedba semielementarnih formula i kontrole te bilo koje enteralne prehrane i totalne parenteralne prehrane. Zaključak: izradom Cochrane sustavnog pregleda i provedbom meta-analize dostupnih literaturnih dokaza prikazani su podatci učinkovitosti i štetnosti primjene različitih oblika enteralne prehrane u bolesnika s akutnim pankreatitisom, s posebnim osvrtom na bolesnike s teškim oblikom bolesti, kao i prednosti, odnosno nedostatci pojedinih nutritivnih terapijskih pristupa. Dodatkom imunomodulatornih tvari enteralnoj prehrani ustanovljeno je značajno smanjenje smrtnosti, ali zbog relativno ograničenog broja uključenih bolesnika potrebna su daljnja istraživanja. Primjena probiotika je kontroverzna i potenijalno povezana s povećanim rizikom razvoja ozbiljnih komplikacija te se njihova rutinska primjena ne može preporučiti. Međutim, zbog heterogenosti i nekonzistentnosti rezultata dodatne strogo kontrolirane i dobro dizajnirane studije su potrebne s ciljem određivanja njihove stvarne štetnosti, odnosno dobrobiti. Semielementarne formule predstavljaju najšire istraživani i primjenjivani oblik enteralne prehrane u akutnom pankreatitisu s jasnom značajnom prednošću u odnosu na prvenstveno totalnu parenteralnu prehranu u poboljšanju kliničkog ishoda i preživljenja bolesnika. Općenito, enteralna prehrana pokazala se učikovitijom u odnosu na totalnu parenteralnu prehranu, kao i pristup bez nutritivne potpore u bolesnika a akutnim pankreatitisom. |
| Abstract (english) | Objectives: this thesis aimed to compare the beneficial and harmful effects of specific enteral nutrition formulations, as well as to compare enteral nutrition to total parenteral nutrition and no nutritional intervention in patients with acute pancreatitis. An additional aim was to evaluate the beneficial and harmful effects of enteral nutrition depending on disease severity. Materials and methods: based on extensive and systematic database searches this systematic review comprises randomized clinical trials comparing a specific type of enteral nutrition formulation to a control consisting of another type of nutritive therapeutical approach (ie, other type of enteral formulation, total parenteral nutrition or no nutritional intervention) in acute pancreatitis. From trials fulfilling the inclusion criteria, characteristics of studies and data regarding primary and secondary outcomes were extracted. Primary outcomes include all-cause mortality, systemic inflammatory response syndrome, organ failure, and adverse events, while secondary outcomes refer to local septic complications, other local complications, other infective complications, C-reactive protein concentrations, and length of hospital stay. The analysis was carried out using Review Manager 5.2, with both the fixed-effect and random-effects model. Results were expressed as risk ratios (RR) for dichotomous data, and mean difference (MD) for continuous data, both with a 95% confidence interval (CI). The analysis was based on an intention-to-treat principle.
Results: searches identified a total of 1308 publications. By eliminating duplicates, as well as non-randomized trials and trials not related to the topic of the review, 41 trials were assessed for eligibility and eventually 25 were included in the systematic review and meta-analysis. Immunonutrition compared to control significantly decreased mortality (RR 0.49; 95% CI 0.29 to 0.80), while no significant difference was shown for other outcomes due to a smaller number of studies included. A similar effect was determined in the subgroup of patients with severe acute pancreatitis. Analysis of enteral nutrition with probiotics showed a significantly lower incidence of organ failure (RR 0.77; 95% CI 0.62 to 0.97) and local septic complications (RR 0.60; 95% CI 0.41 to 0.89), but only by the fixed-effect model due to significant heterogeneity between studies. The occurrence of other infective complications was significantly lower in the probiotics group (RR 0.56; 95% CI 0.32 to 0.98). One study from this analysis found a significantly higher number of serious adverse events, which resulted in higher mortality and number of organ failure. Because of a high risk of bias and possible study design flaws, a post-hoc analysis with exclusion of data from this trial was performed. An additional beneficial effect of probiotics on the above mentioned outcomes was noted. The use of semi-elemental formulas significantly reduced mortality (RR 0.33; 95% CI 0.20 to 0.54), occurrence of organ failure (RR 0.42; 95% CI 0.30 to 0.55) and other local complications (RR 0.48; 95% CI 0.34 to 0.67). Other outcomes were similar in both groups. Efficacy assessment of polymeric enteral formulations was possible for only two trials, with no significant effects regarding all assessed outcomes. Similar results on the majority of clinical outcomes were obtained by the analysis of two trials comparing fiber-enriched enteral nutrition to control, with a significantly shorter length of hospital stay (MD -9.28; 95% CI -13.21 to -5.35) and a lower number of other local complications (11/28 versus 24/32; P=0.008). Only one trial assessed the use of immunonutrition enriched with probiotics and fibers, therefore a meta-analysis was not plausible, showing a significantly shorter length of hospital stay (P=0.01). Subgroup analysis confirmed that any type of enteral nutrition used is significantly more effective than total parenteral nutrition in reduction of mortality (RR 0.33; 95% CI 0.21 to 0.54) and organ failure (RR 0.39; 95% CI 0.30 to 0.52), with a similar number of adverse events. Enteral nutrition has proven better than no nutritional intervention in reduction of mortality (RR 0.50; 95% CI 0.29 to 0.86), while meta-analysis of other primary outcomes was not possible due to lack of data. Trial sequential analysis for meta-analyses of mortality which showed significant difference, confirmed a sufficient amount and power of evidence in case of comparing semi-elemental formulations to control, and any type of enteral nutrition to total parenteral nutrition.
Conclusion: by conducting a Cochrane systematic review and meta-analysis of available literature data, evidence of the beneficial and harmful effects of different enteral nutrition formulations in patients with acute pancreatitis was analyzed. Addition of immunomodulatory agents to enteral nutrition showed significant decrease in mortality, but because of a relatively limited number of patients included further research is required. The use of probiotics is controversial and potentially associated with an increased risk of developement of serious adverse events, therefore their routine application cannot be recommended. However, due to significant heterogeneity and incosistency of results additional rigorously controlled and well-designed trials are needed to determine their potential efficacy or harms. Semi-elemental formulations represent the most widely researched and used type of enteral nutrition in acute pancreatitis with a clear significant benefit over mainly total parenteral nutrition in improving patients clinical outcome and survival. Generally, results showed that enteral nutrition is more effective than total parenteral nutrition, as well as no nutritional support in patients with acute pancreatitis. |
