Postoperative gastrointestinal (GI) dysfunction is one of the most frequent complications in surgical patients. Most cases are associated with episodes of splanchnic hypoperfusion due to hypovolaemia or cardiac dysfunction. It has been suggested that perioperative haemodynamic goal-directed therapy (GDT) may reduce the incidence of these complications in cardiac surgery, and other surgery, but clear evidence is lacking. We have undertaken a meta-analysis of the effects of GDT on postoperative GI and liver complications. A systematic search, using MEDLINE, EMBASE, and The Cochrane Library databases, was performed. Sixteen randomized controlled trials (3410 participants) met the inclusion criteria. Data synthesis was obtained using odds ratio (OR) with 95% confidence interval (CI) by random-effects model. Statistical heterogeneity was assessed by Q and I2 statistics. GI complications were ranked as major (required radiological or surgical intervention or life-threatening condition) or minor (no or only pharmacological treatment required). Major GI complications were significantly reduced by GDT when compared with a control group (OR, 0.42; 95% CI, 0.27–0.65). Minor GI complications were also significantly decreased in the GDT group (OR, 0.29; 95% CI, 0.17–0.50). Treatment did not reduce hepatic injury rate (OR, 0.54; 95% CI, 0.19–1.55). Quality sensitive analyses confirmed the main overall results. In patients undergoing major surgery, GDT, by maintaining an adequate systemic oxygenation, can protect organs particularly at risk of perioperative hypoperfusion and is effective in reducing GI complications.
◊ Reviewed by F. Bonnet
Service d'Anesthésie-Réanimation,
Hôpital Tenon, Paris, France
Major surgery is prone to complications through postoperative gastrointestinal dysfunction. One of the implicated mechanisms could be related to splanchnic hypoperfusion leading to ischaemia or even mesenteric infarction, especially in high-risk patients. Several trials have evaluated the effect of goal-directed haemodynamic therapy to improve oxygen delivery to tissues and to decrease the incidence of gastrointestinal complications. These trials have been included in this systematic review aiming to determine the efficacy of this therapeutic approach.
Randomized controlled trials were eligible if they included patients scheduled for non-cardiac surgery with the main objective to determine the effect of haemodynamic manipulations on postoperative morbidity and mortality. Trials with patients who had previously been admitted to intensive care units for sepsis or shock were excluded from the analysis. The quality of studies was assessed to detect bias. Digestive tract complications were classified as either minor (e.g. diarrhea) or major if these required radiological or surgical intervention and could be life threatening.
Sixteen out of 50 studies retrieved from the literature were included in this meta-analysis based on a total of 3410 patients. Fifteen of these studies were in patients scheduled for abdominal surgery and 11 studies enrolled high-risk patients. The technique of haemodynamic monitoring varied across studies but oesophageal Doppler was the most frequently used technique. The objective of haemodynamic handling also varied among increase in cardiac index, oxygen delivery above a certain threshold, and maintenance of systemic or pulmonary pressure within a certain range. Modalities of optimization were based on the use of fluids and inotropes.
In 11 studies, 106 patients developed major digestive tract complications. The incidence of these major complications was 13% and 6.8% in control and treated groups, respectively (OR: 0.42; 95% CI 0.27-0.65). In 6 studies, 87 patients developed minor complications, of which the incidence was 21% and 8.4% in control and treated groups, respectively (OR: 0.29; 95% CI 0.17-0.50). No difference in liver dysfunction was observed between groups.
Although it seems logical that optimization of haemodynamics during surgery may prevent the occurrence of complications related to a decrease in oxygen tissue perfusion, the results of this meta-analysis deserve several remarks. First of all, there might be bias due to incomplete blinding as in the treated group patients were managed in agreement with cardiac monitoring by oesophageal Doppler which was not used in the control group. In addition, monitoring and objectives of haemodynamic manipulations were different from one study to another. In some studies authors relied on pressure control, in others on cardiac output or oxygen transport or consumption. Eventually, as stressed by the authors, these results are not in line with others demonstrating harm due to increased fluid infusion, especially when considering the incidence of respiratory complications. However, no data, neither on other complications, nor on the duration of hospital stay, or overall mortality have been reported. This will be required to adequately assess the value of goal-directed haemodynamic treatment.
In conclusion, it is likely that some patients may benefit from increasing cardiac output during surgery, in particular those with compromised mesenteric circulation due to atherosclerosis of digestive tract arteries. On the other hand, fluid infusion in excess is reported to prolong the duration of postoperative ileus and may lead to pulmonary complications. Therefore, a more global approach is mandatory to determine the appropriate management of high-risk patients.
|
