|Home | About | Journals | Submit | Contact Us | Français|
A best evidence topic in cardiac surgery was written according to a structured protocol. We address whether routine pharmacological stress ulcer prophylaxis is of benefit for patients undergoing cardiac surgery. One hundred and fifty-six papers were found using the reported search, of which 10 represented the best evidence to answer the clinical question. The authors, journal, date, country of publication, patient group, study type, relevant outcomes and results of these papers were tabulated. The results show that the incidence of stress ulcers following cardiac surgery is low (0.45%), but remains associated with significant morbidity and mortality. Five of the 7 studies demonstrated suppression of acid secretion or decreased incidence of gastric complications in patients given pharmacological stress ulcer prophylaxis, with the remaining two suggesting no clinical benefit. One prospective study of 210 patients, randomized equally between a proton pump inhibitor (PPI), histamine antagonist and teprenone, found that PPIs were the most effective at reducing gastric complications after cardiac surgery, including ulcer formation and upper gastrointestinal bleeding (UGIB). However, a separate retrospective study suggested no difference in the outcomes between the use of a PPI and a histamine antagonist. Of the studies focused on histamine antagonists, one randomized control trial (RCT) showed that cimetidine can reduce surgical stress, augment the immune system and reduce the intubation time after cardiac surgery, although no direct association with UGIB was made. A second prospectively randomized study of histamine antagonists demonstrated superior pH control with famotidine and ranitidine, when compared with cimetidine. Furthermore, haematological and neurological side-effects were noted only with the use of cimetidine. A recent meta-analysis and systematic review of the literature associated gastric acid suppression with an increased risk of pneumonia. Two prospective cohort studies that examined the use of PPI in conjunction with clopidogrel in patients with coronary artery disease concluded that there was no association with an increase in major adverse cardiovascular events with the use of PPIs. We conclude that the current evidence is marginally in favour of the use of prophylactic PPIs. However, this is associated with an increased risk of hospital-acquired pneumonia.
A best evidence topic was constructed according to a structured protocol. This is fully described in the ICVTS .
In [patients undergoing cardiac surgery] are [pharmacological agents used in stress ulcer prophylaxis] effective [in reducing gastrointestinal complications]?
You review a 72-year old lady following coronary artery bypass grafting (CABG). You are concerned that she is developing pneumonia, and your consultant notices you have prescribed lansoprazole. He suggests that the use of PPIs increases the risk of pneumonia with no significant benefit in gastric protection. You decide to look up the evidence to justify your actions.
MedLine was searched from 1948 to November week 1, 2011 using the OVID SP interface. (cardiac surgery.mp OR open heart surgery.mp OR exp cardiac surgical procedures/ OR CABG.mp OR valve surgery.mp) AND (exp omeprazole/ or exp anti-ulcer agents/ or exp proton pump Inhibitors.) A second search (proton pump inhibitors.mp AND pneumonia.mp) limited to review articles was also performed.
One hundred and fifty-six papers were found. Of these, 10 papers provided the best evidence to answer the clinical question (Table 1).
Several studies have examined pharmacological stress ulcer prophylaxis in patients undergoing cardiac surgery. Hata et al.  conducted a prospective randomized trial of antisecretory agents following cardiac surgery. Patients were randomized into three groups of 70: Group 1, teprenone daily; Group 2, ranitidine and Group 3, rabeprazole. Gastric fibroscopy at post-operative days 5 to 7 showed a similar incidence of oesophagitis, superficial gastritis and erosive gastritis. However, the incidence of haemorrhagic gastritis (Group 1: 22.9%; Group 2: 15.7%; Group 3: 2.9%, P = 0.0003) and active ulcers (Group 1: 28.6%; Group 2: 21.4%; Group 3: 4.3%, P = 0.0001) were significantly lower with the PPI use. Haematemesis complications were documented in Groups 1 and 2, but not in Group 3, and gastric pain was also reduced with the PPI use. Gon et al.  reviewed 33 patients following cardiac surgery, prescribed either famotidine or rabeprazole. Peri-operative gastric/oesophageal pH and serological evidence of Helicobacter pylori infection were measured. The results suggested acid suppression by both drugs with no upper gastrointestinal bleeding (UGIB) in either group.
A number of studies have examined the outcomes of histamine antagonists against placebo; Wagner et al.  demonstrated an increase in pH and a decrease in gastric volumes, and Tayama et al.  suggested a reduction in the surgical stress response and augmentation of the immune system with the use of histamine antagonists.
Rosen et al.  performed a retrospective study comparing 32 patients, who developed the life-threatening complication of peptic ulcers after cardiac surgery, with 32 randomly selected controls. Older age, re-exploration and prolonged hypotension were significantly more frequent in patients with ulcers (10 of 32 cases versus 1 of 32 controls; P < 0.005). However, peri-operative ulcer prophylaxis was employed with equal frequency in cases and controls and did not correlate with a significantly different outcome.
Lamothe et al.  compared four agents on their efficacy at preventing ulcers in 57 patients undergoing CABG. No UGIB was recorded, with superior pH control in the famotidine (n = 18) and ranitidine (n = 19) groups (P < 0.003) over antacid (n = 5) or cimetidine (n = 15) groups (pH ≤4.0). Furthermore, haematological and neurological side-effects were noted only in the cimetidine group.
In a literature review by van der Voort et al. , ischaemia, reperfusion injury and endotoxaemia were shown to be the main pathogenic mechanisms in stress ulcer formation. Valve replacement, aortic cross-clamping and bypass time, non-pulsatile flow, re-operation and inflammatory state were risk factors for UGIB. The overall incidence of UGIB in patients after cardiac surgery was low (0.45%) with the incidence in patients on no prophylaxis (0.45%) only slightly higher than those on prophylaxis (0.35%). The authors concluded that the routine use of histamine antagonists is not supported in the available retrospective reports.
Concern remains regarding the risk of pneumonia with acid suppression, numerous studies have yielded inconsistent results. Eom et al.  conducted a systematic review and meta-analysis of acid-suppressive drugs and risk of pneumonia (outside the context of cardiac surgery). Meta-analysis of eight observational studies showed that the risk of pneumonia was higher in those taking PPI (odds ratio [OR] 1.27, 95% confidence interval [CI] 1.11–1.46), and histamine antagonists (OR 1.22; CI 1.09–1.36). Analysis of RCTs showed an increased risk of hospital-acquired pneumonia with acid suppression.
Several studies have investigated the concomitant use of PPIs with clopidogrel, with regards to adverse clinical outcomes in patients with coronary artery disease. Harjai et al.  studied 2651 patients following coronary stenting. All patients received aspirin indefinitely and a thienopyridine for 1–12 months. Patients were categorized into those taking a PPI (n = 751) and those not taking a PPI (n = 1900) at discharge. Primary end points were; 6-month incidence of major adverse cardiovascular events (MACE) (composite of death, myocardial infarction, target vessel revascularization and stent thrombosis) and net adverse clinical events (NACE) (composite of MACE and thrombolysis in myocardial infarction major or minor bleeding), evaluated using propensity-adjusted Cox regression analysis. In addition, propensity-matched analysis was performed in 685 pairs of patients. In propensity-adjusted as well as propensity-matched analyses, the use of PPIs was not associated with an increased risk of MACE or NACE. A similar study by Banerjee et al. , analysing all-cause death, non-fatal myocardial infarction, repeat revascularization and MACE in a cohort of 23 200 post-percutaneous coronary intervention (PCI) patients using a multivariate adjusted Cox model and propensity-matched case–control analysis, came to the same conclusion.
Current evidence is marginally in favour of routine acid-suppression following cardiac surgery for the prophylaxis of gastrointestinal complications. In particular, prophylactic PPIs have been shown to reduce haemorrhagic gastritis and haematemesis complications. Recent meta-analysis suggests an increased risk of pneumonia with routine use of acid-suppression, however, routine use of PPI has not been shown to increase complications related to inhibition of clopidogrel. Individual management decisions must be based on clinical indication. Due to the high morbidity and mortality in patients with gastrointestinal bleeding following cardiac surgery, the authors of this BET would recommend routine acid-suppression with PPI.
Conflict of interest: none declared.