Organ transplantation is the treatment of choice for many end-stage organ diseases. However, it is still strongly limited by organ shortage (1), with increasing disparity between organ supply and demand (2). One promising way to overcome this problem is to optimize brain-dead organ donation.
Brain death is an inflammatory syndrome (3, 4) that causes a massive catecholamine release, with a sudden decrease in cortisol, insulin, thyroid, and pituitary hormone levels (5). Hormonal replacement therapy has been reported to stabilize and improve cardiac function in brain-dead donors (6, 7). Rosendale et al. (8) described a retrospective analysis of more than 10,000 consecutive donors, suggesting that aggressive pharmacologic therapy results in more transplanted organs. However, other studies have failed to confirm these benefits (9, 10).
In addition, hormonal alterations have a profound hemodynamic and metabolic impact on potential donors, inducing a variety of deleterious effects that can threat organ perfusion and cardiac arrest (11). Frequently, hemodynamic collapse precludes organ donation (12). Therefore, early donor management is associated with increased organ retrieval (13), and many transplantation centers and critical care societies have developed standardized donor management protocols that focus on hemodynamic and hormonal resuscitation (14–16).
However, the strategy of brain-dead donor management is still controversial; it is time and resource consuming and has potential deleterious effects, such as metabolic acidosis. Therefore, the present systematic review and meta-analysis was carried out to assess the impact of interventions focusing on the care of brain-dead donors on transplantation outcomes.
Literature Search Results and Study Characteristics
We identified 5094 potentially relevant citations from electronic database search and 2 from manual search. After duplicates were removed, 2726 studies were screened based on title and abstracts, resulting in 39 randomized controlled trials (RCTs) for further evaluation. Twenty studies fulfilled our inclusion criteria, providing data on 1277 patients (Table 1). Of these, 12 trials could not be grouped (single intervention and outcome) (11, 17–27) and were evaluated qualitatively, and 8 studies (15, 16, 28–33) were included in the meta-analysis. A flowchart of search and selection criteria is shown in Figure 1.
The trials were published from 1977 to 2010. The main interventions included desmopressin use (n=146 patients), triiodothyronine replacement (n=264 patients), methylprednisolone replacement (n=150 patients), triiodothyronine and methylprednisolone replacement (n=60 patients), fluid management (n=33 patients), vasopressor therapy (n =264 patients), mechanical ventilation strategies (n=118 patients), and surgical techniques (n=242 patients). The outcomes assessed varied substantially between studies, from hemodynamic parameters and organ function to graft and patient survival. There were no major treatment-associated adverse events.
Table 2 shows the risk of bias in each trial. Eight studies were blinded, 7 used placebo-controlled groups, 9 trials described adequate sequence generation and allocation concealment, 15 used the intention-to-treat principle for statistical analysis, 2 declared to have received grant support from the pharmaceutical industry, and 3 were stopped early (1 because of harm [impaired immediate renal function of kidney recipients], 1 because of benefit, and 1 because of termination of funding).
Trials Included in Qualitative-Only Analysis
Twelve studies (11, 17–27) reported interventions and outcomes not duplicated by other authors. Because they were single studies, it was not possible to group them and they were not therefore included in a meta-analysis. One of them, a well-designed multicenter RCT including 118 patients, demonstrated that more lungs were eligible for transplantation when a lung protective ventilator strategy with low tidal volumes was used compared with a conventional ventilator strategy (24). Another trial, performed in 60 European centers, allocated brain-dead patients to pretreatment with low-dose dopamine or no treatment. The results showed that early kidney graft function was improved with the pharmacologic approach, but with no impact on patient survival (27). Two underpowered studies tested the role of fluid management of donors. When low molecular weight hydroxyethyl starch was administered to liver donors, no differences in early function were found (25). However, impaired immediate renal function in kidney recipients was reported when hydroxyethyl starch was used as plasma expander in kidney donors, leading to early termination of the trial (18).
The impact of methylprednisolone replacement on brain-dead donors was investigated in three different RCTs that evaluated distinct organ outcomes. Hormone therapy was not effective to either increase lung yield (13) or improve early kidney graft function (17). Only one study demonstrated a protective effect of methylprednisolone on liver grafts, with significant down-regulation of inflammation markers and decreased incidence of acute rejection after liver transplantation (23).
Two of six studies of thyroid hormone replacement therapy described study-specific outcomes and were not included in the meta-analysis. The first one measured the impact of triiodothyronine on liver function tests during the first week after transplantation, obtaining similar results in both groups and worse metabolic acidosis in the treatment group (26). The other study showed a trend toward less inotrope need in the treatment group (22).
Circulatory deterioration and cardiac arrest usually occur soon after the diagnosis of brain death. Persistent hypotension despite vasopressor support may be present in up to 20% of donors, especially in hypovolemic patients with untreated diabetes insipidus (5). However, long-term maintenance of circulation of brain-dead donors was reported in a desmopressin-treated group when compared with a control group (21).
No difference regarding graft performance was found in a trial designed to compare two preservation solutions as the initial flush in hepatic allograft procurement (19). Finally, a trial of the impact of donor harvesting technique using a modified double (aortic and portal) perfusion technique for suboptimal liver grafts was terminated early because of benefit, with improved 6-month graft and patient survival rates when compared with a single aortic perfusion technique (20).
Trials Included in Meta-analyses
We retrieved eight studies that could be meta-analyzed: two evaluating desmopressin (n=121 patients), four intravenous triiodothyronine (n=209 patients), and two ischemic liver preconditioning (n=151 patients).
The two studies on desmopressin use (n=121 patients) (30, 31) assessed the effects of desmopressin administration to brain-dead donors on early graft function in kidney recipients. As shown in Figure 2A, no benefits of desmopressin on early graft function of kidney transplants were observed (relative risk [RR], 0.97; 95% confidence interval [CI], 0.85–1.10; I2=0%; P for heterogeneity=0.809).
Four trials (n=209 patients) allocated brain-dead patients to receive intravenous triiodothyronine or placebo and used cardiac index as outcome (9, 10, 32, 33). No differences in cardiac index were found between groups (difference between groups, 0.15; 95% CI, −0.13 to 0.42 L/min/m2; I2=17.4%; P for heterogeneity=0.304; Fig. 2B). Funnel plot analysis did not show significant publication bias for the triiodothyronine intervention.
Figure 2C depicts the meta-analysis of two RCTs (n=151 patients) (28, 29) that assessed the effects of ischemic liver preconditioning during the donor harvesting procedures. No differences were observed in patient survival at 24 to 25 months (RR, 1.00; 95% CI, 0.93–1.08; I2=0%; P for heterogeneity=0.459).
The quality of the evidence was estimated for each of the interventions included in the meta-analysis: desmopressin, intravenous triiodothyronine, and ischemic liver preconditioning. In all cases, the results were based on RCTs, but the quality level of the body of evidence was decreased in two levels due to the high risk of bias in study design and implementation. As the studies demonstrated direct evidence, no statistical heterogeneity of results, narrow 95% CIs, and no publication bias, the overall Grading of Recommendations Assessment, Development and Evaluation (GRADE) quality rating was considered low.
Aggressive donor management protocols must rely on strong pathophysiologic evidence. However, the present systematic review and meta-analyses did not find consistent support for recommending such strategies, especially hormonal replacement.
It has been suggested that the hemodynamic instability associated with brain death is in part a result of diminished levels of circulating thyroxine, leading to a reduction of myocardial energy stores and a shift from aerobic to anaerobic metabolism (7). Experimental studies demonstrated improved cardiac function following thyroid hormonal replacement therapy in brain-dead baboons (34). Many retrospective analyses suggest that thyroid hormonal replacement could improve cardiac function and increase the number of organs transplanted per donor (8, 35, 36). We retrieved six RCTs designed to evaluate the effects of triiodothyronine replacement to organ donors, and all of them had consistent negative results. Moreover, the pooled analysis of four RCTs (n=209 patients) evaluating cardiac index as outcome turned out to be negative. Another recent meta-analysis has evaluated the effect of triiodothyronine replacement on donor heart function, and no benefit was found (37). These inconsistent findings suggest that depletion of triiodothyronine (and the subsequent relative hypothyroid state) is not the major determinant of myocardial dysfunction in these patients but rather perhaps only an adaptive response to illness. Similarly, the evidence in favor of the administration of triiodothyronine in another setting of adaptive relative hypothyroidism, that is, critical care, is far from compelling, to the point that some authors advise withholding its use in critically ill patients unless there is clear evidence of previous hypothyroidism (38).
The complex hemodynamic dysfunction related to brain death is frequently associated with major complications in the potential donor and has multiple causes (1). Diabetes insipidus may be present in up to 80% of these patients, with severe dehydration and hypovolemia (11). The use of desmopressin was not associated with better kidney graft outcomes in the present meta-analysis (30, 31), but it is safe and useful to limit the harmful effects of profuse polyuria, decreasing the need for large volume infusions and preventing hemodynamic collapse (39). Studies have suggested the use of colloid solutions as an option to avoid the infusion of large volumes to treat hypovolemia, because fluid overload could be deleterious to lung grafts (40). The only two RCTs retrieved by us dealing with fluid replacement did not support this notion. On the contrary, one study showed no difference between treatment groups of liver donors when hydroxyethyl starch was used compared with crystalloid solutions. The other study was terminated early because of harm to immediate renal function of kidney recipients (18, 25). To this point, there is no evidence supporting the use of hydroxyethyl starch in brain-dead or other critical care patients (41). Vasodilation and hypotension are almost always present in these patients, but no high-level evidence for the choice of one or another vasopressor agent is available. Donor treatment with dopamine resulted in a reduction in dialysis requirement after kidney transplantation, with no clinically significant impact on graft or patient survival (27). This fact, taken together with the results of a recent meta-analysis of septic shock patients, would advise in favor of norepinephrine, because dopamine was associated with greater mortality and higher incidence of arrhythmias when compared with norepinephrine (42).
Marginal livers, which have been used to increase the donor pool, are especially susceptible to ischemia-reperfusion injury. Ischemic preconditioning during harvesting is a strategy to prevent the deleterious effects of ischemia-reperfusion on the liver graft probably by modulating the inflammatory response (43). Compared with standard orthotopic liver transplantation, this strategy is associated with better tolerance to ischemia but with no significant difference on patient survival (28, 29).
Brain death is associated with a profound proinflammatory process. Under these circumstances, a beneficial role of steroids could be expected, as suggested by retrospective studies (36, 44). However, RCTs using methylprednisolone in brain-dead donors contradict this hypothesis. Methylprednisolone neither increased lung yield (13) nor improved kidney function after transplantation (17). The only positive effect was observed for a surrogate outcome: down-regulation of inflammatory and apoptotic markers in liver biopsies (23). As brain-dead donors might have a relative adrenal insufficiency (45), another potential benefit of corticosteroid use is promotion of hemodynamic stability. However, only methylprednisolone, which lacks significant mineralocorticoid activity, has been evaluated in RCTs. The use of hydrocortisone or even fludrocortisones may result in better outcomes and should be evaluated in future RCTs.
The best evidence in the management of organ donor refers to mechanical ventilation. The use of lung protective strategies with low tidal volumes increases the yield of lungs when compared with conventional ventilatory strategies (24). High tidal volumes are known to be detrimental to patients with acute lung injury (46, 47), and prevention of overdistension seems to be beneficial to potential lung donors.
This study has several limitations. First, there are few RCTs dealing with the management of brain-dead donors. Second, the general quality of the evidence was considered low according to GRADE criteria (48), mostly due to flaws in study design and implementation, raising the possibility of bias. Third, the endpoints of various trials differed, and many of them had evaluated only surrogate hemodynamic endpoints as their primary outcomes, such as hemodynamic parameters or initial organ function. Besides, studies had reported no major treatment-associated adverse events, raising the question if they were properly evaluated. However, similar results were obtained in a systematic review and meta-analysis focusing specifically on clinical trials of thyroid hormone administration to brain-dead potential organ donors (37), which concludes that the use of thyroid hormone in marginal donors is based on low-level evidence.
It should be noted that the interventions found in this study to be ineffective to increase patient or organ survival when used alone are nevertheless recommended by international guidelines (1, 5, 14, 16, 49). However, we believe that a multi-intervention strategy protocol conducted by a dedicated senior physician at the bedside could produce more favorable outcomes. An early combined strategy holding the best choice of fluids, vasopressor drugs, mechanical ventilation parameters, surgical techniques, and combined hormonal replacement therapy should be tested in a well-designed clinical trial.
In summary, despite the implementation of aggressive donor care protocols focusing on hemodynamic and hormonal resuscitation by many transplantation centers and critical care societies, these recommendations are weakly supported, with most evidence based on surrogate outcomes and retrospective data. We recognize the great importance of brain-dead donor care to improve transplantation outcomes, but this systematic review and meta-analysis did not provide consistent evidence for recommending this strategy. Therefore, further RCTs are required to elucidate to what extent a multi-intervention management strategy of brain-dead donors is helpful for transplant recipients.
MATERIALS AND METHODS
Search Strategy and Study Selection
To identify RCTs comparing any category of intervention on management of brain-dead organ donors, in June 2010, we performed an initial electronic literature search in Medline, Embase, and Cochrane as well as the Cochrane Controlled Trials Register, without language or date restriction. This initial search was complemented in August 2012 using the following medical subject headings (MeSH): “Tissue and Organ Procurement” [MeSH] or “Directed Tissue Donation” [MeSH] or “Brain Death” [MeSH]. A high sensitivity strategy for the search of RCTs was used (50). Additionally, we manually searched the references of the selected studies. This systematic review is reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-analyses (51).
We included RCTs comparing interventions aimed to stabilize hemodynamics in brain-dead donors or to improve donated organ function or organ receptor outcomes after transplantation compared with a control group. The following were excluded: (a) RCTs that did not provide information regarding the associations of the intervention with donor stability, organ function, or receptor’s outcomes in the experimental group, the control group, or both, and (b) duplicate publications or substudies of included trials.
Titles and abstracts of retrieved articles were independently evaluated by two reviewers (T.H.R. and R.B.M.). Disagreements were solved by consensus or by a third reviewer (C.B.L.). The investigators were not blinded to authors, institutions, or journals. Articles whose abstracts did not provide enough information regarding the inclusion and exclusion criteria were retrieved for full text evaluation. To avoid possible double counting of patients included in more than one report by the same authors or working groups, recruitment periods were evaluated. If necessary, the corresponding author was contacted for elucidation. Two reviewers (T.H.R. and R.B.M.) independently conducted data extraction.
Assessment of Risk of Bias and Quality of the Evidence
Risk of bias was based on GRADE guidelines (48). Study quality assessment included adequate sequence generation, allocation concealment, blinding of outcomes assessment, and intention-to-treat analysis. Studies without a clear description of an adequate sequence generation or lacking information on allocation concealment criteria were considered not to have fulfilled the inclusion criteria. Quality assessment was independently performed by two reviewers (T.H.R. and R.B.M.) and disagreements solved by consensus or a third reviewer (C.B.L.).
The quality of the evidence was estimated for interventions included in the meta-analysis according to GRADE guidelines and was based on the following: limitations in the study design and implementation, indirectness of evidence, unexplained heterogeneity or inconsistency of results, imprecision of results, and probability of publication bias.
The clinical outcomes of interest were hemodynamic parameters before transplantation, quantification of organ retrieval, organ function after transplantation, and patient or graft survival after transplantation. Studies reporting a similar intervention and outcome were grouped, and a separate forest plot was constructed whenever possible (similar intervention/outcomes in two or more studies identified during the search). For analysis of single studies (that could not be grouped), only qualitative assessment was performed.
For continuous variable outcomes, means or differences between means and respective dispersion values were extracted, and pooled-effect estimates were obtained by comparing the least-squares mean percentage change from baseline to the end of the study for each group; these results were expressed as the weighted mean difference between groups. For categorical outcomes, the total number of patients included and the number of participants with the outcome were used to calculate the overall RR of the intervention to improve an outcome.
Cochran’s Q test was used to evaluate heterogeneity between studies, and a threshold P=0.1 was considered statistically significant. The I2 test was also conducted to evaluate the magnitude of the heterogeneity between studies. We used risk estimates obtained with a fixed-effect meta-analysis because no significant heterogeneity was found between the studies. Publication bias was assessed using a contour-enhanced funnel plot of each trial’s effect size against the standard error (52). Funnel plot asymmetry was evaluated by Begg and Egger tests, and a significant publication bias was considered if P<0.1. The trim-and-fill computation was used to estimate the effect of publication bias on the interpretation of results (53, 54). All statistical analyses were performed by Stata 11.0 software (Stata, College Station, TX).
1. DuBose J, Salim A. Aggressive organ donor management protocol. J Intensive Care Med
2008; 23: 367.
2. Joshi NR, Margulies DR. Aggressive organ donor management: more from less? Curr Opin Organ Transplant
2006; 11: 141.
3. Kusaka M, Pratschke J, Wilhelm MJ, et al.. Activation of inflammatory mediators in rat renal isografts by donor brain death. Transplantation
2000; 69: 405.
4. Nijboer WN, Schuurs TA, van der Hoeven JA, et al.. Effects of brain death on stress and inflammatory response in the human donor kidney. Transplant Proc
2005; 37: 367.
5. Wood KE, Becker BN, McCartney JG, et al.. Care of the potential organ donor. N Engl J Med
2004; 351: 2730.
6. Novitzky D, Cooper DK, Chaffin JS, et al.. Improved cardiac allograft function following triiodothyronine therapy to both donor and recipient. Transplantation
1990; 49: 311.
7. Salim A, Vassiliu P, Velmahos GC, et al.. The role of thyroid hormone administration in potential organ donors. Arch Surg
2001; 136: 1377.
8. Rosendale JD, Kauffman HM, McBride MA, et al.. Aggressive pharmacologic donor management results in more transplanted organs. Transplantation
2003; 75: 482.
9. Goarin JP, Cohen S, Riou B, et al.. The effects of triiodothyronine on hemodynamic status and cardiac function in potential heart donors. Anesth Analg
1996; 83: 41.
10. Mariot J, Jacob F, Voltz C, et al.. Interet de l’hormonotherapie associant triiodothyronine et cortisone chez le patient en etat de mort cerebrale [Value of hormonal treatment with triiodothyronine and cortisone in brain dead patients]. Ann Fr Anesth Reanim
1991; 10: 321.
11. Rech TH, Rodrigues Filho ÉM. Manuseio do potencial doador de múltiplos órgãos. Rev Bras Ter Intensiva
2007; 19: 197.
12. Bugge JF. Brain death and its implications for management of the potential organ donor. Acta Anaesthesiol Scand
2009; 53: 1239.
13. Venkateswaran RV, Patchell VB, Wilson IC, et al.. Early donor management increases the retrieval rate of lungs for transplantation. Ann Thorac Surg
2008; 85: 278; discussion 86.
14. Zaroff JG, Rosengard BR, Armstrong WF, et al.. Consensus conference report: Maximizing use of organs recovered from the cadaver donor: cardiac recommendations, March 28–29, 2001, Crystal City, Va. Circulation
2002; 106: 836.
15. Rosendale JD, Chabalewski FL, McBride MA, et al.. Increased transplanted organs from the use of a standardized donor management protocol. Am J Transplant
2002; 2: 761.
16. Westphal GA, Caldeira Filho M, Vieira KD, et al.. Guidelines for potential multiple organ donors (adult). Part II. Mechanical ventilation, endocrine metabolic management, hematological and infectious aspects. Rev Bras Ter Intensiva
2001; 23: 269.
17. Chatterjee SN, Terasaki PI, Fine S, et al.. Pretreatment of cadaver donors with methylprednisolone in human renal allografts. Surg Gynecol Obstet
1977; 145: 729.
18. Cittanova ML, Leblanc I, Legendre C, et al.. Effect of hydroxyethylstarch in brain-dead kidney donors on renal function in kidney-transplant recipients. Lancet
1996; 348: 1620.
19. Cofer JB, Klintmalm GB, Morris CV, et al.. A prospective randomized trial between Euro-Collins and University of Wisconsin solutions as the initial flush in hepatic allograft procurement. Transplantation
1992; 53: 995.
20. D’Amico F, Vitale A, Gringeri E, et al.. Liver transplantation using suboptimal grafts: impact of donor harvesting technique. Liver Transpl
2007; 13: 1444.
21. Iwai A, Sakano T, Uenishi M, et al.. Effects of vasopressin and catecholamines on the maintenance of circulatory stability in brain-dead patients. Transplantation
1989; 48: 613.
22. Jeevanandam V. Triiodothyronine: spectrum of use in heart transplantation. Thyroid
1997; 7: 139.
23. Kotsch K, Ulrich F, Reutzel-Selke A, et al.. Methylprednisolone therapy in deceased donors reduces inflammation in the donor liver and improves outcome after liver transplantation: a prospective randomized controlled trial. Ann Surg
2008; 248: 1042.
24. Mascia L, Pasero D, Slutsky AS, et al.. Effect of a lung protective strategy for organ donors on eligibility and availability of lungs for transplantation: a randomized controlled trial. JAMA
2010; 304: 2620.
25. Randell T, Orko R, Hockerstedt K. Peroperative fluid management of the brain-dead multiorgan donor. Acta Anaesthesiol Scand
1990; 34: 592.
26. Randell TT, Hockerstedt KA. Triiodothyronine treatment in brain-dead multiorgan donors—a controlled study. Transplantation
1992; 54: 736.
27. Schnuelle P, Gottmann U, Hoeger S, et al.. Effects of donor pretreatment with dopamine on graft function after kidney transplantation: a randomized controlled trial. JAMA
2009; 302: 1067.
28. Amador A, Grande L, Marti J, et al.. Ischemic pre-conditioning in deceased donor liver transplantation: a prospective randomized clinical trial. Am J Transplant
2007; 7: 2180.
29. Azoulay D, Del Gaudio M, Andreani P, et al.. Effects of 10 minutes of ischemic preconditioning of the cadaveric liver on the graft’s preservation and function: the ying and the yang. Ann Surg
2005; 242: 133.
30. Guesde R, Barrou B, Leblanc I, et al.. Administration of desmopressin in brain-dead donors and renal function in kidney recipients. Lancet
1998; 352: 1178.
31. Pennefather SH, Bullock RE, Mantle D, et al.. Use of low dose arginine vasopressin to support brain-dead organ donors. Transplantation
1995; 59: 58.
32. Perez-Blanco A, Caturla-Such J, Canovas-Robles J, et al.. Efficiency of triiodothyronine treatment on organ donor hemodynamic management and adenine nucleotide concentration. Intensive Care Med
2005; 31: 943.
33. Venkateswaran RV, Steeds RP, Quinn DW, et al.. The haemodynamic effects of adjunctive hormone therapy in potential heart donors: a prospective randomized double-blind factorially designed controlled trial. Eur Heart J
2009; 30: 1771.
34. Novitzky D, Wicomb WN, Cooper DKC, et al.. Electrocardiographic, hemodynamic and endocrine changes occurring during experimental brain death in the chacma baboon. J Heart Transplant
1984; 4: 63.
35. Novitzky D, Cooper DK, Reichart B. Hemodynamic and metabolic responses to hormonal therapy in brain-dead potential organ donors. Transplantation
1987; 43: 852.
36. Rosendale JD, Kauffman HM, McBride MA, et al.. Hormonal resuscitation yields more transplanted hearts, with improved early function. Transplantation
2003; 75: 1336.
37. Macdonald PS, Aneman A, Bhonagiri D, et al.. A systematic review and meta-analysis of clinical trials of thyroid hormone administration to brain dead potential organ donors. Crit Care Med
2012; 40: 1635.
38. Stathatos N, Levetan C, Burman KD, et al.. The controversy of the treatment of critically ill patients with thyroid hormone. Best Pract Res Clin Endocrinol Metab
2001; 15: 465.
39. Dictus C, Vienenkoetter B, Esmaeilzadeh M, et al.. Critical care management of potential organ donors: our current standard. Clin Transplant
2009; 23: 2.
40. Pennefather SH, Bullock RE, Dark JH. The effect of fluid therapy on alveolar arterial oxygen gradient in brain-dead organ donors. Transplantation
1993; 56: 1418.
41. Gattas DJ, Dan A, Myburgh J, et al.. Fluid resuscitation with 6% hydroxyethyl starch (130/0.4) in acutely ill patients: an updated systematic review and meta-analysis. Anesth Analg
2012; 114: 159.
42. De Backer D, Aldecoa C, Njimi H, et al.. Dopamine versus norepinephrine in the treatment of septic shock: a meta-analysis. Crit Care Med
2012; 40: 725.
43. Jassem W, Fuggle SV, Cerundolo L, et al.. Ischemic preconditioning of cadaver donor livers protects allografts following transplantation. Transplantation
2006; 81: 169.
44. Follette DM, Rudich SM, Babcock WD. Improved oxygenation and increased lung donor recovery with high-dose steroid administration after brain death. J Heart Lung Transplant
1998; 17: 423.
45. Dimopoulou I, Tsagarakis S, Anthi A, et al.. High prevalence of decreased cortisol reserve in brain-dead potential organ donors. Crit Care Med
2003; 31: 1113.
46. Amato MB, Barbas CS, Medeiros DM, et al.. Effect of a protective-ventilation strategy on mortality in the acute respiratory distress syndrome. N Engl J Med
1998; 338: 347.
47. Brower RG, Lanken PN, MacIntyre N, et al.. Higher versus lower positive end-expiratory pressures in patients with the acute respiratory distress syndrome. N Engl J Med
2004; 351: 327.
48. Balshem H, Helfand M, Schunemann HJ, et al.. GRADE guidelines: 3. Rating the quality of evidence. J Clin Epidemiol
2011; 64: 401.
49. Powner DJ, Darby JM, Kellum JA. Proposed treatment guidelines for donor care. Prog Transplant
2004; 14: 16; quiz 7–8.
50. Robinson KA, Dickersin K. Development of a highly sensitive search strategy for the retrieval of reports of controlled trials using PubMed. Int J Epidemiol
2002; 31: 150.
51. Moher D, Liberati A, Tetzlaff J, et al.. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Ann Intern Med
2009; 151: 264.
52. Peters JL, Sutton AJ, Jones DR, et al.. Contour-enhanced meta-analysis funnel plots help distinguish publication bias from other causes of asymmetry. J Clin Epidemiol
2008; 61: 991.
53. Higgins JP, Thompson SG, Deeks JJ, et al.. Measuring inconsistency in meta-analyses. BMJ
2003; 327: 557.
54. Duval S, Tweedie R. Trim and fill: a simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysis. Biometrics
2000; 56: 455.