Systematic reviews and meta-analyses of randomized trials: principles and pitfalls

Systematic reviews and meta-analyses allow for a more transparent and objective appraisal of the evidence. They may decrease the number of false-negative results and prevent delays in the introduction of effective interventions into clinical practice. However, as for any other tool, their misuse can result in severely misleading results. In this article, we discuss the main steps that should be taken when conducting systematic reviews and meta-analyses, namely the preparation of a review protocol, identification of eligible trials, and data extraction, pooling of treatment effects across trials, investigation of potential reasons for differences in treatment effects across trials, and complete reporting of the review methods and findings. We also discuss common pitfalls that should be avoided, including the use of quality assessment tools to derive summary quality scores, pooling of data across trials as if they belonged to a single large trial, and inappropriate uses of meta-regression that could result in misleading estimates of treatment effects because of regression to the mean or the ecological fallacy. If conducted and reported properly, systematic reviews and meta-analyses will increase our understanding of the strengths and weaknesses of the available evidence, which may eventually facilitate clinical decision making.

In vitro Activity of Fosfomycin Alone and in Combination with Ceftriaxone or Azithromycin Against Clinical Neisseria gonorrhoeae Isolates.

New therapeutic strategies are needed to combat the emergence of infections due to multidrug-resistant Neisseria gonorrhoeae (Ng). In this study, fosfomycin (FOS) was tested against 89 Ng using the Etest method and showing MIC50/90s of only 8/16 μg/ml (range ≤ 1-32 μg/ml). FOS in combination with ceftriaxone (CRO) or azithromycin (AZT) was then evaluated using the checkerboard method for eight strains, including F89 (CRO-resistant) and AZT-HLR (high-level AZT-resistant). All combinations including FOS gave indifferent effects (fractional inhibitory concentration [FIC] index values between 1.2-2.3 for FOS plus CRO and between 1.8-3.2 for FOS plus AZT). Time-kill experiments for FOS, CRO, AZT and their combinations (at concentrations of 0.5×, 1×, 2× and 4× MIC) were performed against ATCC 49226, one Ng of NG-MAST ST1407, F89 and AZT-HLR. For all strains, at 24 hours results indicated that: i) FOS was bactericidal at 2× MIC concentrations but after >24 hours there was re-growth of bacteria; ii) CRO was bactericidal at 0.5× MIC; iii) AZT was bactericidal at 4× MIC; iv) CRO plus AZT was less bactericidal than CRO alone; v) FOS plus AZT was bactericidal at 2× MIC; vi) CRO plus AZT and FOS plus CRO were both bactericidal at 0.5× MIC, but the latter had more rapid effects. FOS is appealing for the management of Ng infections because of its single and oral formulation. However, our results suggest its use in combination with CRO. This strategy could, after appropriate clinical trials, be implemented for the treatment of infections due to isolates possessing resistance to CRO and/or AZT.

Prognostic implications of coronary calcification in patients with obstructive coronary artery disease treated by percutaneous coronary intervention: a patient-level pooled analysis of 7 contemporary stent trials

OBJECTIVE To investigate the long-term prognostic implications of coronary calcification in patients undergoing percutaneous coronary intervention for obstructive coronary artery disease. METHODS Patient-level data from 6296 patients enrolled in seven clinical drug-eluting stents trials were analysed to identify in angiographic images the presence of severe coronary calcification by an independent academic research organisation (Cardialysis, Rotterdam, The Netherlands). Clinical outcomes at 3-years follow-up including all-cause mortality, death-myocardial infarction (MI), and the composite end-point of all-cause death-MI-any revascularisation were compared between patients with and without severe calcification. RESULTS Severe calcification was detected in 20% of the studied population. Patients with severe lesion calcification were less likely to have undergone complete revascularisation (48% vs 55.6%, p<0.001) and had an increased mortality compared with those without severely calcified arteries (10.8% vs 4.4%, p<0.001). The event rate was also high in patients with severely calcified lesions for the combined end-point death-MI (22.9% vs 10.9%; p<0.001) and death-MI- any revascularisation (31.8% vs 22.4%; p<0.001). On multivariate Cox regression analysis, including the Syntax score, the presence of severe coronary calcification was an independent predictor of poor prognosis (HR: 1.33 95% CI 1.00 to 1.77, p=0.047 for death; 1.23, 95% CI 1.02 to 1.49, p=0.031 for death-MI, and 1.18, 95% CI 1.01 to 1.39, p=0.042 for death-MI- any revascularisation), but it was not associated with an increased risk of stent thrombosis. CONCLUSIONS Patients with severely calcified lesions have worse clinical outcomes compared to those without severe coronary calcification. Severe coronary calcification appears as an independent predictor of worse prognosis, and should be considered as a marker of advanced atherosclerosis.

Comparative therapeutic value of post-remission approaches in patients with acute myeloid leukemia aged 40-60 years.

The preferred type of post-remission therapy (PRT) in patients with acute myeloid leukemia (AML) in first complete remission (CR1) is a subject of continued debate, especially in patients at higher risk of nonrelapse mortality (NRM), including patients >40 years of age. We report results of a time-dependent multivariable analysis of allogenic hematopoietic stem cell transplantation (alloHSCT) (n=337) versus chemotherapy (n=271) or autologous HSCT (autoHSCT) (n=152) in 760 patients aged 40-60 years with AML in CR1. Patients receiving alloHSCT showed improved overall survival (OS) as compared with chemotherapy (respectively, 57±3% vs 40±3% at 5 years, P<0.001). Comparable OS was observed following alloHSCT and autoHSCT in patients with intermediate-risk AML (60±4 vs 54±5%). However, alloHSCT was associated with less relapse (hazard ratio (HR) 0.51, P<0.001) and better relapse-free survival (RFS) (HR 0.74, P=0.029) as compared with autoHSCT in intermediate-risk AMLs. AlloHSCT was applied following myeloablative conditioning (n=157) or reduced intensity conditioning (n=180), resulting in less NRM, but comparable outcome with respect to OS, RFS and relapse. Collectively, these results show that alloHSCT is to be preferred over chemotherapy as PRT in patients with intermediate- and poor-risk AML aged 40-60 years, whereas autoHSCT remains a treatment option to be considered in patients with intermediate-risk AML.Leukemia advance online publication, 23 December 2014; doi:10.1038/leu.2014.332.

Endovascular treatment for acute ischemic stroke patients: implications and interpretation of IMS III, MR RESCUE, and SYNTHESIS EXPANSION trials: A report from the Working Group of International Congress of Interventional Neurology

OBJECTIVE The results of Interventional Management of Stroke (IMS) III, Magnetic Resonance and REcanalization of Stroke Clots Using Embolectomy (MR RESCUE), and SYNTHESIS EXPANSION trials are expected to affect the practice of endovascular treatment for acute ischemic stroke. The purpose of this report is to review the components of the designs and methods of these trials and to describe the influence of those components on the interpretation of trial results. METHODS A critical review of trial design and conduct of IMS III, MR RESCUE, and SYNTHESIS EXPANSION is performed with emphasis on patient selection, shortcomings in procedural aspects, and methodology of data ascertainment and analysis. The influence of each component is estimated based on published literature including multicenter clinical trials reporting on endovascular treatment for acute ischemic stroke and myocardial infarction. RESULTS We critically examined the time interval between symptom onset and treatment and rates of angiographic recanalization to differentiate between "endovascular treatment" and "parameter optimized endovascular treatment" as it relates to the IMS III, MR RESCUE, and SYNTHESIS EXPANSION trials. All the three trials failed to effectively test "parameter optimized endovascular treatment" due to the delay between symptom onset and treatment and less than optimal rates of recanalization. In all the three trials, the magnitude of benefit with endovascular treatment required to reject the null hypothesis was larger than could be expected based on previous studies. The IMS III and SYNTHESIS EXPANSION trials demonstrated that rates of symptomatic intracerebral hemorrhages subsequent to treatment are similar between IV thrombolytics and endovascular treatment in matched acute ischemic stroke patients. The trials also indirectly validated the superiority/equivalence of IV thrombolytics (compared with endovascular treatment) in patients with minor neurological deficits and those without large vessel occlusion on computed tomographic/magnetic resonance angiography. CONCLUSIONS The results do not support a large magnitude benefit of endovascular treatment in subjects randomized in all the three trials. The possibility that benefits of a smaller magnitude exist in certain patient populations cannot be excluded. Large magnitude benefits can be expected with implementation of "parameter optimized endovascular treatment" in patients with ischemic stroke who are candidates for IV thrombolytics.

Response to neoadjuvant treatment of invasive ductal breast carcinomas including outcome evaluation: MRI analysis by an automatic CAD system in comparison to visual evaluation

BACKGROUND The aim of this study was to evaluate imaging-based response to standardized neoadjuvant chemotherapy (NACT) regimen by dynamic contrast-enhanced magnetic resonance mammography (DCE-MRM), whereas MR images were analyzed by an automatic computer-assisted diagnosis (CAD) system in comparison to visual evaluation. MRI findings were correlated with histopathologic response to NACT and also with the occurrence of metastases in a follow-up analysis. PATIENTS AND METHODS Fifty-four patients with invasive ductal breast carcinomas received two identical MRI examinations (before and after NACT; 1.5T, contrast medium gadoteric acid). Pre-therapeutic images were compared with post-therapeutic examinations by CAD and two blinded human observers, considering morphologic and dynamic MRI parameters as well as tumor size measurements. Imaging-assessed response to NACT was compared with histopathologically verified response. All clinical, histopathologic, and DCE-MRM parameters were correlated with the occurrence of distant metastases. RESULTS Initial and post-initial dynamic parameters significantly changed between pre- and post-therapeutic DCE-MRM. Visually evaluated DCE-MRM revealed sensitivity of 85.7%, specificity of 91.7%, and diagnostic accuracy of 87.0% in evaluating the response to NACT compared to histopathology. CAD analysis led to more false-negative findings (37.0%) compared to visual evaluation (11.1%), resulting in sensitivity of 52.4%, specificity of 100.0%, and diagnostic accuracy of 63.0%. The following dynamic MRI parameters showed significant associations to occurring metastases: Post-initial curve type before NACT (entire lesions, calculated by CAD) and post-initial curve type of the most enhancing tumor parts after NACT (calculated by CAD and manually). CONCLUSIONS In the accurate evaluation of response to neoadjuvant treatment, CAD systems can provide useful additional information due to the high specificity; however, they cannot replace visual imaging evaluation. Besides traditional prognostic factors, contrast medium-induced dynamic MRI parameters reveal significant associations to patient outcome, i.e. occurrence of distant metastases.

Preclinical studies identify non-apoptotic low-level caspase-3 as therapeutic target in pemphigus vulgaris

The majority of pemphigus vulgaris (PV) patients suffer from a live-threatening loss of intercellular adhesion between keratinocytes (acantholysis). The disease is caused by auto-antibodies that bind to desmosomal cadherins desmoglein (Dsg) 3 or Dsg3 and Dsg1 in mucous membranes and skin. A currently unresolved controversy in PV is whether apoptosis is involved in the pathogenic process. The objective of this study was to perform preclinical studies to investigate apoptotic pathway activation in PV pathogenesis with the goal to assess its potential for clinical therapy. For this purpose, we investigated mouse and human skin keratinocyte cultures treated with PV antibodies (the experimental Dsg3 monospecific antibody AK23 or PV patients IgG), PV mouse models (passive transfer of AK23 or PVIgG into adult and neonatal mice) as well as PV patients' biopsies (n=6). A combination of TUNEL assay, analyses of membrane integrity, early apoptotic markers such as cleaved poly-ADP-ribose polymerase (PARP) and the collapse of actin cytoskeleton failed to provide evidence for apoptosis in PV pathogenesis. However, the in vitro and in vivo PV models, allowing to monitor progression of lesion formation, revealed an early, transient and low-level caspase-3 activation. Pharmacological inhibition confirmed the functional implication of caspase-3 in major events in PV such as shedding of Dsg3, keratin retraction, proliferation including c-Myc induction, p38MAPK activation and acantholysis. Together, these data identify low-level caspase-3 activation downstream of disrupted Dsg3 trans- or cis-adhesion as a major event in PV pathogenesis that is non-synonymous with apoptosis and represents, unlike apoptotic components, a promising target for clinical therapy. At a broader level, these results posit that an impairment of adhesive functions in concert with low-level, non-lethal caspase-3 activation can evoke profound cellular changes which may be of relevance for other diseases including cancer.

Consensus on the Need for a Hierarchical List of Patient-reported Pain Outcomes for Metaanalyses of Knee Osteoarthritis Trials: An OMERACT Objective.

OBJECTIVE Although protocol registration for systematic reviews is still not mandatory, reviewers should be strongly encouraged to register the protocol to identify the methodological approach, including all outcomes of interest. This will minimize the likelihood of biased decisions in reviews, such as selective outcome reporting. A group of international experts convened to address issues regarding the need to develop hierarchical lists of outcome measurement instruments for a particular outcome for metaanalyses. METHODS Multiple outcome measurement instruments exist to measure the same outcome. Metaanalysis of knee osteoarthritis (OA) trials, and the assessment of pain as an outcome, was used as an exemplar to assess how Outcome Measures in Rheumatology (OMERACT), the Cochrane Collaboration, and other international initiatives might contribute in this area. The meeting began with formal presentations of background topics, empirical evidence from the literature, and a brief introduction to 2 existing hierarchical lists of pain outcome measurement instruments recommended for metaanalyses of knee OA trials. RESULTS After discussions, most participants agreed that there is a need to develop a methodology for generation of hierarchical lists of outcome measurement instruments to guide metaanalyses. Tools that could be used to steer development of such a prioritized list are the COSMIN checklist (Consensus-based Standards for the selection of health status Measurement Instruments) and the OMERACT Filter 2.0. CONCLUSION We list meta-epidemiological research agenda items that address the frequency of reported outcomes in trials, as well as methodologies to assess the best measurement properties (i.e., truth, discrimination, and feasibility).

Taking personalized medicine seriously: Biomarker approaches in phase IIb/III studies in major depression and schizophrenia

The success rate in the development of psychopharmacological compounds is insufficient. Two main reasons for failure have been frequently identified: 1) treating the wrong patients and 2) using the wrong dose. This is potentially based on the known heterogeneity among patients, both on a syndromal and a biological level. A focus on personalized medicine through better characterization with biomarkers has been successful in other therapeutic areas. Nevertheless, obstacles toward this goal that exist are 1) the perception of a lack of validation, 2) the perception of an expensive and complicated enterprise, and 3) the perception of regulatory hurdles. The authors tackle these concerns and focus on the utilization of biomarkers as predictive markers for treatment outcome. The authors primarily cover examples from the areas of major depression and schizophrenia. Methodologies covered include salivary and plasma collection of neuroendocrine, metabolic, and inflammatory markers, which identified subgroups of patients in the Netherlands Study of Depression and Anxiety. A battery of vegetative markers, including sleep-electroencephalography parameters, heart rate variability, and bedside functional tests, can be utilized to characterize the activity of a functional system that is related to treatment refractoriness in depression (e.g., the renin-angiotensin-aldosterone system). Actigraphy and skin conductance can be utilized to classify patients with schizophrenia and provide objective readouts for vegetative activation as a functional marker of target engagement. Genetic markers, related to folate metabolism, or folate itself, has prognostic value for the treatment response in patients with schizophrenia. Already, several biomarkers are routinely collected in standard clinical trials (e.g., blood pressure and plasma electrolytes), and appear to be differentiating factors for treatment outcome. Given the availability of a wide variety of markers, the further development and integration of such markers into clinical research is both required and feasible in order to meet the benefit of personalized medicine. This article is based on proceedings from the "Taking Personalized Medicine Seriously-Biomarker Approaches in Phase IIb/III Studies in Major Depression and Schizophrenia" session, which was held during the 10th Annual Scientific Meeting of the International Society for Clinical Trials Meeting (ISCTM) in Washington, DC, February 18 to 20, 2014.

Differentiation between Staphylococcus aureus and coagulase-negative Staphylococcus species by real-time PCR including detection of methicillin resistants in comparison to conventional microbiology testing.

BACKGROUND Staphylococcus aureus has long been recognized as a major pathogen. Methicillin-resistant strains of S. aureus (MRSA) and methicillin-resistant strains of S. epidermidis (MRSE) are among the most prevalent multiresistant pathogens worldwide, frequently causing nosocomial and community-acquired infections. METHODS In the present pilot study, we tested a polymerase chain reaction (PCR) method to quickly differentiate Staphylococci and identify the mecA gene in a clinical setting. RESULTS Compared to the conventional microbiology testing the real-time PCR assay had a higher detection rate for both S. aureus and coagulase-negative Staphylococci (CoNS; 55 vs. 32 for S. aureus and 63 vs. 24 for CoNS). Hands-on time preparing DNA, carrying out the PCR, and evaluating results was less than 5 h. CONCLUSIONS The assay is largely automated, easy to adapt, and has been shown to be rapid and reliable. Fast detection and differentiation of S. aureus, CoNS, and the mecA gene by means of this real-time PCR protocol may help expedite therapeutic decision-making and enable earlier adequate antibiotic treatment.

Long-term results of therapeutic local anesthesia (neural therapy) in 280 referred refractory chronic pain patients

BACKGROUND: Can the application of local anesthetics (Neural Therapy, NT) alone durably improve pain symptoms in referred patients with chronic and refractory pain? If the application of local anesthetics does lead to an improvement that far exceeds the duration of action of local anesthetics, we will postulate that a vicious circle of pain in the reflex arcs has been disrupted (hypothesis). METHODS: Case series design. We exclusively used procaine or lidocaine. The inclusion criteria were severe pain and chronic duration of more than three months, pain unresponsive to conventional medical measures, written referral from physicians or doctors of chiropractic explicitly to NT. Patients with improvement of pain who started on additional therapy during the study period for a reason other than pain were excluded in order to avoid a potential bias. Treatment success was measured after one year follow-up using the outcome measures of pain and analgesics intake. RESULTS: 280 chronic pain patients were included; the most common reason for referral was back pain. The average number of consultations per patient was 9.2 in the first year (median 8.0). After one year, in 60 patients pain was unchanged, 52 patients reported a slight improvement, 126 were considerably better, and 41 pain-free. At the same time, 74.1 % of the patients who took analgesics before starting NT needed less or no more analgesics at all. No adverse effects or complications were observed. CONCLUSIONS: The good long-term results of the targeted therapeutic local anesthesia (NT) in the most problematic group of chronic pain patients (unresponsive to all evidence based conventional treatment options) indicate that a vicious circle has been broken. The specific contribution of the intervention to these results cannot be determined. The low costs of local anesthetics, the small number of consultations needed, the reduced intake of analgesics, and the lack of adverse effects also suggest the practicality and cost-effectiveness of this kind of treatment. Controlled trials to evaluate the true effect of NT are needed.

Is dosing of therapeutic immunoglobulins optimal? A review of a three-decade long debate in europe.

The consumption of immunoglobulins (Ig) is increasing due to better recognition of antibody deficiencies, an aging population, and new indications. This review aims to examine the various dosing regimens and research developments in the established and in some of the relevant off-label indications in Europe. The background to the current regulatory settings in Europe is provided as a backdrop for the latest developments in primary and secondary immunodeficiencies and in immunomodulatory indications. In these heterogeneous areas, clinical trials encompassing different routes of administration, varying intervals, and infusion rates are paving the way toward more individualized therapy regimens. In primary antibody deficiencies, adjustments in dosing and intervals will depend on the clinical presentation, effective IgG trough levels and IgG metabolism. Ideally, individual pharmacokinetic profiles in conjunction with the clinical phenotype could lead to highly tailored treatment. In practice, incremental dosage increases are necessary to titrate the optimal dose for more severely ill patients. Higher intravenous doses in these patients also have beneficial immunomodulatory effects beyond mere IgG replacement. Better understanding of the pharmacokinetics of Ig therapy is leading to a move away from simplistic "per kg" dosing. Defective antibody production is common in many secondary immunodeficiencies irrespective of whether the causative factor was lymphoid malignancies (established indications), certain autoimmune disorders, immunosuppressive agents, or biologics. This antibody failure, as shown by test immunization, may be amenable to treatment with replacement Ig therapy. In certain immunomodulatory settings [e.g., idiopathic thrombocytopenic purpura (ITP)], selection of patients for Ig therapy may be enhanced by relevant biomarkers in order to exclude non-responders and thus obtain higher response rates. In this review, the developments in dosing of therapeutic immunoglobulins have been limited to high and some medium priority indications such as ITP, Kawasaki' disease, Guillain-Barré syndrome, chronic inflammatory demyelinating polyradiculoneuropathy, myasthenia gravis, multifocal motor neuropathy, fetal alloimmune thrombocytopenia, fetal hemolytic anemia, and dermatological diseases.

Current developments in the use of stem cell for therapeutic neovascularisation: is the future therapy "cell-free"?

The plasticity and self-regenerative properties of stem cells have opened new avenues in regenerative medicine. Greater understanding of the biology of stem cells is followed by growing expectations of a rapid translation into alternative therapeutic options. Recent preclinical studies and clinical trials employing stem and progenitor cells from different sources have shown encouraging results. However, their underlying mechanisms are still poorly understood, the potential adverse effects and the discrepancy in efficacy remain to be further investigated. Their essential role in vessel regeneration has made endothelial progenitor cells (EPC) a suitable candidate for therapeutic applications aiming at tissue revascularisation. Recent evidence suggests that EPC contribute to neovascularisation not only by direct participation in tissue homeostasis but mainly via paracrine mechanisms. In future, novel therapeutic strategies could be based on EPC paracrine factors or synthetic factors, and replace cell transplantation.

What is the influence of randomisation sequence generation and allocation concealment on treatment effects of physical therapy trials? A meta-epidemiological study.

OBJECTIVE To determine if adequacy of randomisation and allocation concealment is associated with changes in effect sizes (ES) when comparing physical therapy (PT) trials with and without these methodological characteristics. DESIGN Meta-epidemiological study. PARTICIPANTS A random sample of randomised controlled trials (RCTs) included in meta-analyses in the PT discipline were identified. INTERVENTION Data extraction including assessments of random sequence generation and allocation concealment was conducted independently by two reviewers. To determine the association between sequence generation, and allocation concealment and ES, a two-level analysis was conducted using a meta-meta-analytic approach. PRIMARY AND SECONDARY OUTCOME MEASURES association between random sequence generation and allocation concealment and ES in PT trials. RESULTS 393 trials included in 43 meta-analyses, analysing 44 622 patients contributed to this study. Adequate random sequence generation and appropriate allocation concealment were accomplished in only 39.7% and 11.5% of PT trials, respectively. Although trials with inappropriate allocation concealment tended to have an overestimate treatment effect when compared with trials with adequate concealment of allocation, the difference was non-statistically significant (ES=0.12; 95% CI -0.06 to 0.30). When pooling our results with those of Nuesch et al, we obtained a pooled statistically significant value (ES=0.14; 95% CI 0.02 to 0.26). There was no difference in ES in trials with appropriate or inappropriate random sequence generation (ES=0.02; 95% CI -0.12 to 0.15). CONCLUSIONS Our results suggest that when evaluating risk of bias of primary RCTs in PT area, systematic reviewers and clinicians implementing research into practice should pay attention to these biases since they could exaggerate treatment effects. Systematic reviewers should perform sensitivity analysis including trials with low risk of bias in these domains as primary analysis and/or in combination with less restrictive analyses. Authors and editors should make sure that allocation concealment and random sequence generation are properly reported in trial reports.

Dexmedetomidine versus standard care sedation with propofol or midazolam in intensive care: an economic evaluation.

INTRODUCTION Dexmedetomidine was shown in two European randomized double-blind double-dummy trials (PRODEX and MIDEX) to be non-inferior to propofol and midazolam in maintaining target sedation levels in mechanically ventilated intensive care unit (ICU) patients. Additionally, dexmedetomidine shortened the time to extubation versus both standard sedatives, suggesting that it may reduce ICU resource needs and thus lower ICU costs. Considering resource utilization data from these two trials, we performed a secondary, cost-minimization analysis assessing the economics of dexmedetomidine versus standard care sedation. METHODS The total ICU costs associated with each study sedative were calculated on the basis of total study sedative consumption and the number of days patients remained intubated, required non-invasive ventilation, or required ICU care without mechanical ventilation. The daily unit costs for these three consecutive ICU periods were set to decline toward discharge, reflecting the observed reduction in mean daily Therapeutic Intervention Scoring System (TISS) points between the periods. A number of additional sensitivity analyses were performed, including one in which the total ICU costs were based on the cumulative sum of daily TISS points over the ICU period, and two further scenarios, with declining direct variable daily costs only. RESULTS Based on pooled data from both trials, sedation with dexmedetomidine resulted in lower total ICU costs than using the standard sedatives, with a difference of €2,656 in the median (interquartile range) total ICU costs-€11,864 (€7,070 to €23,457) versus €14,520 (€7,871 to €26,254)-and €1,649 in the mean total ICU costs. The median (mean) total ICU costs with dexmedetomidine compared with those of propofol or midazolam were €1,292 (€747) and €3,573 (€2,536) lower, respectively. The result was robust, indicating lower costs with dexmedetomidine in all sensitivity analyses, including those in which only direct variable ICU costs were considered. The likelihood of dexmedetomidine resulting in lower total ICU costs compared with pooled standard care was 91.0% (72.4% versus propofol and 98.0% versus midazolam). CONCLUSIONS From an economic point of view, dexmedetomidine appears to be a preferable option compared with standard sedatives for providing light to moderate ICU sedation exceeding 24 hours. The savings potential results primarily from shorter time to extubation. TRIAL REGISTRATION ClinicalTrials.gov NCT00479661 (PRODEX), NCT00481312 (MIDEX).