Perioperative glucocorticoid administration for prevention of systemic organ failure in patients undergoing esophageal resection for esophageal carcinoma

ABSTRACT CONTEXT AND OBJECTIVE: Preoperative glucocorticoid administration has been proposed for reducing postoperative morbidity. This is not widely used before esophageal resection because of incomplete knowledge regarding its effectiveness. The aim here was to assess the effects of preoperative glucocorticoid administration in adults undergoing esophageal resection for esophageal carcinoma. SEARCH STRATEGY: Studies were identified by searching the Cochrane Controlled Trials Register, MEDLINE, EMBASE, Cancer Lit, SCIELO and Cochrane Library, and by manual searching from relevant articles. The last search for clinical trials for this systematic review was performed in December 2004. SELECTION CRITERIA: This review included randomized studies of patients with potentially resectable carcinomas of the esophagus that compared preoperative glucocorticoid administration with placebo. DATA COLLECTION AND ANALYSIS: Data were extracted by the same reviewers, and the trial quality was assessed using Jadad scoring. Relative risk and weighted mean difference with 95% confidence limits were used to assess the significance of the difference between the treatment arms. RESULTS: Four randomized trials involving 146 patients were found. There were no differences in postoperative mortality, sepsis, anastomotic leakage, hepatic and renal failure between the glucocorticoid and placebo groups. There were fewer postoperative respiratory complications (p = 0.005) and multiple postoperative complications (p = 0.004) and lower postoperative plasma interleukin-6 levels (p = 0.00001) with preoperative glucocorticoid administration. There was a higher postoperative PaO2/FiO2 ratio (p = 0.0001) with preoperative glucocorticoid administration. CONCLUSION: Prophylactic administration of glucocorticoids is associated with decreased postoperative complications.

DATA COLLECTION AND ANALYSIS: Data were extracted by the same reviewers, and the trial quality was assessed using Jadad scoring.Relative risk and weighted mean difference with 95% confidence limits were used to assess the significance of the difference between the treatment arms.
RESULTS: Four randomized trials involving 146 patients were found.There were no differences in postoperative mortality, sepsis, anastomotic leakage, hepatic and renal failure between the glucocorticoid and placebo groups.There were fewer postoperative respiratory complications (p = 0.005) and multiple postoperative complications (p = 0.004) and lower postoperative plasma interleukin-6 levels (p = 0.00001) with preoperative glucocorticoid administration.There was a higher postoperative PaO 2 /FiO 2 ratio (p = 0.0001) with preoperative glucocorticoid administration.
CONCLUSION: Prophylactic administration of glucocorticoids is associated with decreased postoperative complications.
compared preoperative glucocorticoid administration with placebo.Studies were identified by searching the Cochrane Controlled Trials Register, Medline (1966 -2004), Embase (1988 -2004), CancerLit (1993 -2004), SciELO (1993 -2004) and Cochrane Library.In addition, manual searching from relevant articles and personal files was included.We also contacted manufacturers and researchers.There were no language restrictions.Data were extracted by the same reviewers, and the trial quality was assessed using Jadad scoring. 12The inclusion of a study depended on the evaluation of the randomization.The most important criterion for the classification was the allocation concealment, which needed to have been maintained until the time of the intervention. 13The data were collected using software from the Cochrane Collaboration: Review Manager, Version 4.2.3 for Windows, Oxford (UK).
The primary outcome of interest was overall mortality.Other variables examined were: anastomotic leakage, postoperative respiratory complications, renal failure, hepatic failure, multiple postoperative complications (two or more complications per patient), postoperative plasma levels of interleukin-6, postoperative PaO 2 /FiO 2 ratio and postoperative hospital stay.
Sensitivity analysis was planned in order to explore sources of heterogeneity, where heterogeneity existed.The factors hypothesized a priori included quality of study and biologically effective corticosteroid dose.Relative risk (RR) and weighted mean difference (WMD) with 95% confidence limits were used to assess the significance of the difference between the treatment arms. 13,14

RESULTS
Four randomized controlled clinical trials comparing glucocorticoid (methylpredniso-lone) with placebo were found in the literature search.Thus, 146 patients were divided into two groups: the placebo group consisted of 77 patients who received injections of saline solution, and the intervention group consisted of 69 patients to whom methylprednisolone was administered intravenously, before the induction of anesthesia.The characteristics of the clinical trials included in the systematic review are shown in Table 1.
There were no differences in postoperative mortality, anastomotic leakage and postoperative hospital stay between the glucocorticoid and placebo groups.
A summary of the meta-analysis results for each variable is presented in Tables 2 and  3, with the numbers of studies included, the numbers of participants, and the results from the heterogeneity and overall effect tests.
The results from the meta-analysis of mean PaO 2 /FiO 2 and respiratory complications are shown in Figures 1 and 2, respectively.

DISCUSSION
There is as yet no agreement on the beneficial effects of corticosteroids in alleviating surgical stress. 1 This disagreement probably stems from the variability in the drugs used, their dosage and administration schedules, and the nature of the surgical procedures in different studies.
5][6][7] Esophageal cancer surgery was therefore selected as one of the most suitable procedures for evaluating the effects of steroids on surgical stress.
Modification of the inflammatory response at an early stage would seem to be very important, because compensatory antiinflammatory responses occur in very quick succession after the inflammatory response.In these studies, therefore, methylprednisolone was administered just before the surgery.
While the anti-inflammatory actions of methylprednisolone are five times as strong as those of cortisol, the actions on electrolyte metabolism are less than half as strong. 11,15,16revious studies on humans and rabbits have demonstrated that methylprednisolone appears in the lung in greater concentrations than prednisolone. 17Also, the half-life of methylprednisolone in the blood is 2.8 hours.When methylprednisolone is administered intravenously at a individual dose of 1,000 mg, the maximum blood concentration in healthy adults is about 10 µg/ml, and more than 10 µg/ml. 11ostoperative mortality was the same for the methylprednisolone and placebo groups in this meta-analysis.The trials included were relatively small and might not have had sufficient statistical power to detect a clinically significant difference in mortality between the groups.
Administration of methylprednisolone reduced the incidence of multiple and respiratory complications and increased the PaO 2 /FiO 2 after the first postoperative day.
The reduction in multiple complications may have been due in part to a decrease in respiratory complications.Furthermore, the rate of anastomotic leaks was similar in the methylprednisolone and placebo groups.These results suggest that preoperative methylprednisolone administration may be safe for alleviating surgical stress.
-The pathophysiology of postoperative organ dysfunction is multifactorial, including additional factors such as hypoxemia, nutrition, pain, type of analgesia, immobilization and surgical expertise. 6,7,18The incidence of sepsis and hepatic and renal failures were similar between the methylprednisolone and placebo groups.Nevertheless, multiple postoperative complications were reduced by administering methylprednisolone before surgery.The reason for this apparent discrepancy is the magnification effect that occurs when multiple complications are gathered together.
The present study also showed that

Relative risk (RR) is the ratio of respiratory complications between the treatment group and the control group; CI = confidence interval; df = degrees of freedom.
preoperative methylprednisolone administration suppressed postoperative increases in the plasma levels of interleukin-6 (IL-6).IL-6 produced by endothelial cells influences the permeability of cultured endothelial cells to albumin, and anti-IL-6 antibodies can prevent increased endothelial permeability.Studies on heart surgery have suggested that increased levels of proinflammatory cytokines, in particular IL-6, can be correlated with impaired hemodynamics and higher incidence of postoperative complications. 3,8,9Therefore, it should be noted that decreased IL-6 levels may be more important for systemic inflammatory responses than for local reactions.

CI
of perioperative inflammatory responses and organ dysfunction.However, these potential beneficial effects and risks from preoperative methylprednisolone administration should be assessed in large-scale studies.
day; WMD = weighted mean difference; CI = confidence interval; df = degrees of freedom.

Figure 2 .
Figure 2. Postoperative respiratory complications in four studies included in this meta-analysis.

CONCLUSIONS
The present meta-analysis has shown that the prophylactic administration of methylprednisolone decreased the numbers of postoperative respiratory complications and multiple postoperative complications and the postoperative plasma levels of IL-6.Methylprednisolone pretreatment represents a potentially important biological modifier Sao Paulo Med J. 2006;124(2):112-5.

PaO 2 /
FiO 2 ratio = mean postoperative ratio (kPa/FiO 2 ) -1 kPa = 7.6 mmHg; WMD = weighted mean difference.The weight given to each study is determined by the precision of its estimated effect and is equal to the inverse of the variance; CI = confidence interval; df = degrees of freedom.

Figure 1 .
Figure 1.Mean postoperative PaO 2 /FiO 2 ratio in two studies included in this meta-analysis.

Table 1 .
Clinical trials included in this systematic review and meta-analysis * Methylprednisolone: 10 mg/kg before induction of anesthesia; † Methylprednisolone: 30 mg/kg before induction of anesthesia.

Table 2 .
Discrete data: summary of this meta-analysis results

Table 3 .
Continuous data: summary of the meta-analysis results