[Evaluation of the "Schema-focused Emotive Behavioural Therapy" (SET) for Patients with Personality Disorders: Results of a Randomised Controlled Trial.]

The "Schema-focussed Emotive Behavioral Therapy" (SET) was developed by our research group as a new group therapy approach for patients with personality disorders from all clusters (A to C; DSM-IV). It was evaluated in a randomised controlled study (n = 93). Data were collected before and after treatment as well as one year after study entry. A completer analysis was conducted with matched subgroups (n = 60). After therapy, SET patients improved in the outcome domains interactional behavior, strain, and symptomatic complaints (IIP-D, GAF, VEV-VW, BSI-P). Furthermore, they showed a significant lower dropout rate. At the follow-up assessment, Cluster C patients of the experimental group deteriorated with regard to symptomatic complaints (BSI-P). In contrast, cluster B patients improved more over time compared to control subjects. SET seems to be an adequate and effective group therapy with effects that seem to be stable over time, especially for patients with Cluster B diagnosis.

Effect of atazanavir versus other protease inhibitor-containing antiretroviral therapy on endothelial function in HIV-infected persons: randomised controlled trial

OBJECTIVE: Impaired endothelial function was demonstrated in HIV-infected persons on protease inhibitor (PI)-containing antiretroviral therapy, probably due to altered lipid metabolism. Atazanavir is a PI causing less atherogenic lipoprotein changes. This study determined whether endothelial function improves after switching from other PI to atazanavir. DESIGN: Randomised, observer-blind, treatment-controlled trial. SETTING: Three university-based outpatient clinics. PATIENTS: 39 HIV-infected persons with suppressed viral replication on PI-containing regimens and fasting low-density lipoprotein (LDL)-cholesterol greater than 3 mmol/l. INTERVENTION: Patients were randomly assigned to continue the current PI or change to unboosted atazanavir. MAIN OUTCOME MEASURES: Endpoints at week 24 were endothelial function assessed by flow-mediated dilation (FMD) of the brachial artery, lipid profiles and serum inflammation and oxidative stress parameters. RESULTS: Baseline characteristics and mean FMD values of the two treatment groups were comparable (3.9% (SD 1.8) on atazanavir versus 4.0% (SD 1.5) in controls). After 24 weeks' treatment, FMD decreased to 3.3% (SD 1.4) and 3.4% (SD 1.7), respectively (all p = ns). Total cholesterol improved in both groups (p<0.0001 and p = 0.01, respectively) but changes were more pronounced on atazanavir (p = 0.05, changes between groups). High-density lipoprotein and triglyceride levels improved on atazanavir (p = 0.03 and p = 0.003, respectively) but not in controls. Serum inflammatory and oxidative stress parameters did not change; oxidised LDL improved significantly in the atazanavir group. CONCLUSIONS: The switch from another PI to atazanavir in treatment-experienced patients did not result in improvement of endothelial function despite significantly improved serum lipids. Atherogenic lipid profiles and direct effects of antiretroviral drugs on the endothelium may affect vascular function. Trial registration number: NCT00447070.

Biolimus-eluting stent with biodegradable polymer versus sirolimus-eluting stent with durable polymer for coronary revascularisation (LEADERS): a randomised non-inferiority trial

BACKGROUND: A novel stent platform eluting biolimus, a sirolimus analogue, from a biodegradable polymer showed promising results in preliminary studies. We compared the safety and efficacy of a biolimus-eluting stent (with biodegradable polymer) with a sirolimus-eluting stent (with durable polymer). METHODS: We undertook a multicentre, assessor-blind, non-inferiority study in ten European centres. 1707 patients aged 18 years or older with chronic stable coronary artery disease or acute coronary syndromes were centrally randomised by a computer-generated allocation sequence to treatment with either biolimus-eluting (n=857) or sirolimus-eluting (n=850) stents. The primary endpoint was a composite of cardiac death, myocardial infarction, or clinically-indicated target vessel revascularisation within 9 months. Analysis was by intention to treat. 427 patients were randomly allocated to angiographic follow-up, with in-stent percentage diameter stenosis as principal outcome measure at 9 months. The trial is registered with ClinicalTrials.gov, number NCT00389220. FINDINGS: We analysed all randomised patients. Biolimus-eluting stents were non-inferior to sirolimus-eluting stents for the primary endpoint at 9 months (79 [9%] patients vs 89 [11%], rate ratio 0.88 [95% CI 0.64-1.19], p for non-inferiority=0.003, p for superiority=0.39). Frequency of cardiac death (14 [1.6%] vs 21 [2.5%], p for superiority=0.22), myocardial infarction (49 [5.7%] vs 39 [4.6%], p=0.30), and clinically-indicated target vessel revascularisation (38 [4.4%] vs 47 [5.5%], p=0.29) were similar for both stent types. 168 (79%) patients in the biolimus-eluting group and 167 (78%) in the sirolimus-eluting group had data for angiographic follow-up available. Biolimus-eluting stents were non-inferior to sirolimus-eluting stents in in-stent percentage diameter stenosis (20.9%vs 23.3%, difference -2.2% [95% CI -6.0 to 1.6], p for non-inferiority=0.001, p for superiority=0.26). INTERPRETATION: Our results suggest that a stent eluting biolimus from a biodegradable polymer represents a safe and effective alternative to a stent eluting sirolimus from a durable polymer in patients with chronic stable coronary artery disease or acute coronary syndromes. FUNDING: Biosensors Europe SA, Switzerland.

Fuel metabolism during exercise in euglycaemia and hyperglycaemia in patients with type 1 diabetes mellitus--a prospective single-blinded randomised crossover trial

We assessed systemic and local muscle fuel metabolism during aerobic exercise in patients with type 1 diabetes at euglycaemia and hyperglycaemia with identical insulin levels.

A randomised controlled trial of spinal manipulative therapy in acute low back pain

OBJECTIVE: To determine whether treatment with spinal manipulative therapy (SMT) administered in addition to standard care is associated with clinically relevant early reductions in pain and analgesic consumption. METHODS: 104 patients with acute low back pain were randomly assigned to SMT in addition to standard care (n = 52) or standard care alone (n = 52). Standard care consisted of general advice and paracetamol, diclofenac or dihydrocodeine as required. Other analgesic drugs or non-pharmacological treatments were not allowed. Primary outcomes were pain intensity assessed on the 11-point box scale (BS-11) and analgesic use based on diclofenac equivalence doses during days 1-14. An extended follow-up was performed at 6 months. RESULTS: Pain reductions were similar in experimental and control groups, with the lower limit of the 95% CI excluding a relevant benefit of SMT (difference 0.5 on the BS-11, 95% CI -0.2 to 1.2, p = 0.13). Analgesic consumptions were also similar (difference -18 mg diclofenac equivalents, 95% CI -43 mg to 7 mg, p = 0.17), with small initial differences diminishing over time. There were no differences between groups in any of the secondary outcomes and stratified analyses provided no evidence for potential benefits of SMT in specific patient groups. The extended follow-up showed similar patterns. CONCLUSIONS: SMT is unlikely to result in relevant early pain reduction in patients with acute low back pain.

Promotion of health in older people: a randomised controlled trial of health risk appraisal in British general practice

BACKGROUND: there is inadequate evidence to support currently formulated NHS strategies to achieve health promotion and preventative care in older people through broad-based screening and assessment in primary care. The most extensively evaluated delivery instrument for this purpose is Health Risk Appraisal (HRA). This article describes a trial using HRA to evaluate the effect on health behaviour and preventative-care uptake in older people in NHS primary care. METHODS: a randomised controlled trial was undertaken in three London primary care group practices. Functionally independent community-dwelling patients older than 65 years (n = 2,503) received a self-administered Health Risk Appraisal for Older Persons (HRA-O) questionnaire leading to computer-generated individualised written feedback to participants and general practitioners (GPs), integrated into practice information-technology (IT) systems. All primary care staff received training in preventative health in older people. The main outcome measures were self-reported health behaviour and preventative care uptake at 1-year follow-up. RESULTS: of 2,503 individuals randomised, 2,006 respondents (80.1%) (intervention, n = 940, control n = 1,066) were available for analysis. Intervention group respondents reported slightly higher pneumococcal vaccination uptake and equivocal improvement in physical activity levels compared with controls. No significant differences were observed for any other categories of health behaviour or preventative care measures at 1-year follow-up. CONCLUSIONS: HRA-O implemented in this way resulted in minimal improvement of health behaviour or uptake of preventative care measures in older people. Supplementary reinforcement involving contact by health professionals with patients over and above routine clinical encounters may be a prerequisite to the effectiveness of IT-based delivery systems for health promotion in older people.

Liraglutide once a day versus exenatide twice a day for type 2 diabetes: a 26-week randomised, parallel-group, multinational, open-label trial (LEAD-6)

BACKGROUND: Unlike most antihyperglycaemic drugs, glucagon-like peptide-1 (GLP-1) receptor agonists have a glucose-dependent action and promote weight loss. We compared the efficacy and safety of liraglutide, a human GLP-1 analogue, with exenatide, an exendin-based GLP-1 receptor agonist. METHODS: Adults with inadequately controlled type 2 diabetes on maximally tolerated doses of metformin, sulphonylurea, or both, were stratified by previous oral antidiabetic therapy and randomly assigned to receive additional liraglutide 1.8 mg once a day (n=233) or exenatide 10 microg twice a day (n=231) in a 26-week open-label, parallel-group, multinational (15 countries) study. The primary outcome was change in glycosylated haemoglobin (HbA(1c)). Efficacy analyses were by intention to treat. The trial is registered with ClinicalTrials.gov, number NCT00518882. FINDINGS: Mean baseline HbA(1c) for the study population was 8.2%. Liraglutide reduced mean HbA(1c) significantly more than did exenatide (-1.12% [SE 0.08] vs -0.79% [0.08]; estimated treatment difference -0.33; 95% CI -0.47 to -0.18; p<0.0001) and more patients achieved a HbA(1c) value of less than 7% (54%vs 43%, respectively; odds ratio 2.02; 95% CI 1.31 to 3.11; p=0.0015). Liraglutide reduced mean fasting plasma glucose more than did exenatide (-1.61 mmol/L [SE 0.20] vs -0.60 mmol/L [0.20]; estimated treatment difference -1.01 mmol/L; 95% CI -1.37 to -0.65; p<0.0001) but postprandial glucose control was less effective after breakfast and dinner. Both drugs promoted similar weight losses (liraglutide -3.24 kg vs exenatide -2.87 kg). Both drugs were well tolerated, but nausea was less persistent (estimated treatment rate ratio 0.448, p<0.0001) and minor hypoglycaemia less frequent with liraglutide than with exenatide (1.93 vs 2.60 events per patient per year; rate ratio 0.55; 95% CI 0.34 to 0.88; p=0.0131; 25.5%vs 33.6% had minor hypoglycaemia). Two patients taking both exenatide and a sulphonylurea had a major hypoglycaemic episode. INTERPRETATION: Liraglutide once a day provided significantly greater improvements in glycaemic control than did exenatide twice a day, and was generally better tolerated. The results suggest that liraglutide might be a treatment option for type 2 diabetes, especially when weight loss and risk of hypoglycaemia are major considerations.

Hands-on time during cardiopulmonary resuscitation is affected by the process of teambuilding: A prospective randomised simulator-based trial

Background Cardiac arrests are handled by teams rather than by individual health-care workers. Recent investigations demonstrate that adherence to CPR guidelines can be less than optimal, that deviations from treatment algorithms are associated with lower survival rates, and that deficits in performance are associated with shortcomings in the process of team-building. The aim of this study was to explore and quantify the effects of ad-hoc team-building on the adherence to the algorithms of CPR among two types of physicians that play an important role as first responders during CPR: general practitioners and hospital physicians. Methods To unmask team-building this prospective randomised study compared the performance of preformed teams, i.e. teams that had undergone their process of team-building prior to the onset of a cardiac arrest, with that of teams that had to form ad-hoc during the cardiac arrest. 50 teams consisting of three general practitioners each and 50 teams consisting of three hospital physicians each, were randomised to two different versions of a simulated witnessed cardiac arrest: the arrest occurred either in the presence of only one physician while the remaining two physicians were summoned to help ("ad-hoc"), or it occurred in the presence of all three physicians ("preformed"). All scenarios were videotaped and performance was analysed post-hoc by two independent observers. Results Compared to preformed teams, ad-hoc forming teams had less hands-on time during the first 180 seconds of the arrest (93 ± 37 vs. 124 ± 33 sec, P < 0.0001), delayed their first defibrillation (67 ± 42 vs. 107 ± 46 sec, P < 0.0001), and made less leadership statements (15 ± 5 vs. 21 ± 6, P < 0.0001). Conclusion Hands-on time and time to defibrillation, two performance markers of CPR with a proven relevance for medical outcome, are negatively affected by shortcomings in the process of ad-hoc team-building and particularly deficits in leadership. Team-building has thus to be regarded as an additional task imposed on teams forming ad-hoc during CPR. All physicians should be aware that early structuring of the own team is a prerequisite for timely and effective execution of CPR.

Ultrasound-guided pudendal nerve block in cats undergoing perineal urethrostomy: a prospective, randomised, investigator-blind, placebo-controlled clinical trial

The objective of this study was to evaluate the clinical usefulness, in terms of analgesic efficacy and safety, of ultrasound-guided pudendal nerve block performed with bupivacaine in cats undergoing perineal urethrostomy. Eighteen client-owned male cats scheduled for perineal urethrostomy were enrolled in the study and assigned to one of two treatment groups. The pudendal nerve block was performed under general anaesthesia as described elsewhere, with 0.3 ml/kg of either saline (group C) or 0.5% bupivacaine (group B) - the total injection volume being split equally on the two sites of injection (left and right). Intra-operatively, assessment of nociception was based on the rescue analgesics requirement, as well as on the evaluation of changes in physiological parameters in comparison with the baseline values. Post-operative pain assessment was performed using three different pain scales at recovery and then 1, 2 and 3 h after recovery. Cats in group B showed lower heart rates and required fewer analgesics during surgery than group C. Post-operatively, group B had lower pain scores and needed less rescue buprenorphine than group C. Iatrogenic block-related complications were not observed. In conclusion, the ultrasound-guided pudendal nerve block can be considered clinically useful in feline medicine as it provides reliable analgesia in cats undergoing perineal urethrostomy.

Effect of a chitosan additive to a Sn(2+)-containing toothpaste on its anti-erosive/anti-abrasive efficacy-a controlled randomised in situ trial

OBJECTIVES It is well known that Sn(2+) is a notable anti-erosive agent. There are indications that biopolymers such as chitosan can enhance the effect of Sn(2+), at least in vitro. However, little information exists about their anti-erosive/anti-abrasive in situ effects. In the present in situ study, the efficacy of Sn(2+)-containing toothpastes in the presence or absence of chitosan was tested. METHODS Ten subjects participated in the randomised crossover study, wearing mandibular appliances with human enamel specimens. Specimens were extraorally demineralised (7 days, 0.5 % citric acid, pH 2.6; 6 × 2 min/day) and intraorally exposed to toothpaste suspensions (2 × 2 min/day). Within the suspension immersion time, one half of the specimens were additionally brushed intraorally with a powered toothbrush (5 s, 2.5 N). Tested preparations were a placebo toothpaste (negative control), two experimental toothpastes (F/Sn = 1,400 ppm F(-), 3,500 ppm Sn(2+); F/Sn/chitosan = 1,400 ppm F(-), 3,500 ppm Sn(2+), 0.5 % chitosan) and an SnF2-containing gel (positive control, GelKam = 3,000 ppm Sn(2+), 1,000 ppm F(-)). Substance loss was quantified profilometrically (μm). RESULTS In the placebo group, tissue loss was 11.2 ± 4.6 (immersion in suspension) and 17.7 ± 4.7 (immersion in suspension + brushing). Immersion in each Sn(2+)-containing suspension significantly reduced tissue loss (p ≤ 0.01); after immersion in suspension + brushing, only the treatments with GelKam (5.4 ± 5.5) and with F/Sn/chitosan (9.6 ± 5.6) significantly reduced loss [both p ≤ 0.05 compared to placebo; F/Sn 12.8 ± 6.4 (not significant)] CONCLUSION Chitosan enhanced the efficacy of the Sn(2+)-containing toothpaste as an anti-erosive/anti-abrasive agent. CLINICAL RELEVANCE The use of Sn(2+)- and chitosan-containing toothpaste is a good option for symptomatic therapy in patients with regular acid impacts.

Effect of comprehensive cardiac rehabilitation after heart valve surgery (CopenHeartVR): study protocol for a randomised clinical trial

Background Heart valve diseases are common with an estimated prevalence of 2.5% in the Western world. The number is rising due to an ageing population. Once symptomatic, heart valve diseases are potentially lethal, and heavily influence daily living and quality of life. Surgical treatment, either valve replacement or repair, remains the treatment of choice. However, post surgery, the transition to daily living may become a physical, mental and social challenge. We hypothesise that a comprehensive cardiac rehabilitation programme can improve physical capacity and self-assessed mental health and reduce hospitalisation and healthcare costs after heart valve surgery. Methods A randomised clinical trial, CopenHeartVR, aims to investigate whether cardiac rehabilitation in addition to usual care is superior to treatment as usual after heart valve surgery. The trial will randomly allocate 210 patients, 1:1 intervention to control group, using central randomisation, and blinded outcome assessment and statistical analyses. The intervention consists of 12 weeks of physical exercise, and a psycho-educational intervention comprising five consultations. Primary outcome is peak oxygen uptake (VO2 peak) measured by cardiopulmonary exercise testing with ventilatory gas analysis. Secondary outcome is self-assessed mental health measured by the standardised questionnaire Short Form 36. Also, long-term healthcare utilisation and mortality as well as biochemistry, echocardiography and cost-benefit will be assessed. A mixed-method design is used to evaluate qualitative and quantitative findings encompassing a survey-based study before the trial and a qualitative pre- and post-intervention study. Discussion The study is approved by the local regional Research Ethics Committee (H-1-2011-157), and the Danish Data Protection Agency (j.nr. 2007-58-0015).

Vandetanib in locally advanced or metastatic differentiated thyroid cancer: a randomised, double-blind, phase 2 trial

BACKGROUND No effective standard treatment exists for patients with radioiodine-refractory, advanced differentiated thyroid carcinoma. We aimed to assess efficacy and safety of vandetanib, a tyrosine kinase inhibitor of RET, VEGFR and EGFR signalling, in this setting. METHODS In this randomised, double-blind, phase 2 trial, we enrolled adults (aged ≥18 years) with locally advanced or metastatic differentiated thyroid carcinoma (papillary, follicular, or poorly differentiated) at 16 European medical centres. Eligible patients were sequentially randomised in a 1:1 ratio with a standard computerised scheme to receive either vandetanib 300 mg per day (vandetanib group) or matched placebo (placebo group), balanced by centre. The primary endpoint was progression-free survival (PFS) in the intention-to-treat population based on investigator assessment. This study is registered with ClinicalTrials.gov, number NCT00537095. FINDINGS Between Sept 28, 2007, and Oct 16, 2008, we randomly allocated 72 patients to the vandetanib group and 73 patients to the placebo group. By data cutoff (Dec 2, 2009), 113 (78%) patients had progressed (52 [72%] patients in the vandetanib group and 61 [84%] in the placebo group) and 40 (28%) had died (19 [26%] patients in the vandetanib group and 21 [29%] in the placebo group). Patients who received vandetanib had longer PFS than did those who received placebo (hazard ratio [HR] 0·63, 60% CI 0·54-0·74; one-sided p=0·008): median PFS was 11·1 months (95% CI 7·7-14·0) for patients in the vandetanib group and 5·9 months (4·0-8·9) for patients in the placebo group. The most common grade 3 or worse adverse events were QTc prolongation (ten [14%] of 73 patients in the vandetanib group vs none in the placebo group), diarrhoea (seven [10%] vs none), asthenia (five [7%] vs three [4%]), and fatigue (four [5%] vs none). Two patients in the vandetanib group and one in the placebo group died from treatment-related serious adverse events (haemorrhage from skin metastases and pneumonia in the vandetanib group and pneumonia in the placebo group). INTERPRETATION Vandetanib is the first targeted drug to show evidence of efficacy in a randomised phase 2 trial in patients with locally advanced or metastatic differentiated thyroid carcinoma. Further investigation of tyrosine-kinase inhibitors in this setting is warranted. FUNDING AstraZeneca.

2 years versus 1 year of adjuvant trastuzumab for HER2-positive breast cancer (HERA): an open-label, randomised controlled trial

BACKGROUND Trastuzumab has established efficacy against breast cancer with overexpression or amplification of the HER2 oncogene. The standard of care is 1 year of adjuvant trastuzumab, but the optimum duration of treatment is unknown. We compared 2 years of treatment with trastuzumab with 1 year of treatment, and updated the comparison of 1 year of trastuzumab versus observation at a median follow-up of 8 years, for patients enrolled in the HERceptin Adjuvant (HERA) trial. METHODS The HERA trial is an international, multicentre, randomised, open-label, phase 3 trial comparing treatment with trastuzumab for 1 and 2 years with observation after standard neoadjuvant chemotherapy, adjuvant chemotherapy, or both in 5102 patients with HER2-positive early breast cancer. The primary endpoint was disease-free survival. The comparison of 2 years versus 1 year of trastuzumab treatment involved a landmark analysis of 3105 patients who were disease-free 12 months after randomisation to one of the trastuzumab groups, and was planned after observing at least 725 disease-free survival events. The updated intention-to-treat comparison of 1 year trastuzumab treatment versus observation alone in 3399 patients at a median follow-up of 8 years (range 0-10) is also reported. This study is registered with ClinicalTrials.gov, number NCT00045032. FINDINGS We recorded 367 events of disease-free survival in 1552 patients in the 1 year group and 367 events in 1553 patients in the 2 year group (hazard ratio [HR] 0·99, 95% CI 0·85-1·14, p=0·86). Grade 3-4 adverse events and decreases in left ventricular ejection fraction during treatment were reported more frequently in the 2 year treatment group than in the 1 year group (342 [20·4%] vs 275 [16·3%] grade 3-4 adverse events, and 120 [7·2%] vs 69 [4·1%] decreases in left ventricular ejection fraction, respectively). HRs for a comparison of 1 year of trastuzumab treatment versus observation were 0·76 (95% CI 0·67-0·86, p<0·0001) for disease-free survival and 0·76 (0·65-0·88, p=0·0005) for overall survival, despite crossover of 884 (52%) patients from the observation group to trastuzumab therapy. INTERPRETATION 2 years of adjuvant trastuzumab is not more effective than is 1 year of treatment for patients with HER2-positive early breast cancer. 1 year of treatment provides a significant disease-free and overall survival benefit compared with observation and remains the standard of care. FUNDING F Hoffmann-La Roche (Roche).

Adjuvant bevacizumab-containing therapy in triple-negative breast cancer (BEATRICE): primary results of a randomised, phase 3 trial

BACKGROUND The addition of bevacizumab to chemotherapy improves progression-free survival in metastatic breast cancer and pathological complete response rates in the neoadjuvant setting. Micrometastases are dependent on angiogenesis, suggesting that patients might benefit from anti-angiogenic strategies in the adjuvant setting. We therefore assessed the addition of bevacizumab to chemotherapy in the adjuvant setting for women with triple-negative breast cancer. METHODS For this open-label, randomised phase 3 trial we recruited patients with centrally confirmed triple-negative operable primary invasive breast cancer from 360 sites in 37 countries. We randomly allocated patients aged 18 years or older (1:1 with block randomisation; stratified by nodal status, chemotherapy [with an anthracycline, taxane, or both], hormone receptor status [negative vs low], and type of surgery) to receive a minimum of four cycles of chemotherapy either alone or with bevacizumab (equivalent of 5 mg/kg every week for 1 year). The primary endpoint was invasive disease-free survival (IDFS). Efficacy analyses were based on the intention-to-treat population, safety analyses were done on all patients who received at least one dose of study drug, and plasma biomarker analyses were done on all treated patients consenting to biomarker analyses and providing a measurable baseline plasma sample. This trial is registered with ClinicalTrials.gov, number NCT00528567. FINDINGS Between Dec 3, 2007, and March 8, 2010, we randomly assigned 1290 patients to receive chemotherapy alone and 1301 to receive bevacizumab plus chemotherapy. Most patients received anthracycline-containing therapy; 1638 (63%) of the 2591 patients had node-negative disease. At the time of analysis of IDFS, median follow-up was 31·5 months (IQR 25·6-36·8) in the chemotherapy-alone group and 32·0 months (27·5-36·9) in the bevacizumab group. At the time of the primary analysis, IDFS events had been reported in 205 patients (16%) in the chemotherapy-alone group and in 188 patients (14%) in the bevacizumab group (hazard ratio [HR] in stratified log-rank analysis 0·87, 95% CI 0·72-1·07; p=0·18). 3-year IDFS was 82·7% (95% CI 80·5-85·0) with chemotherapy alone and 83·7% (81·4-86·0) with bevacizumab and chemotherapy. After 200 deaths, no difference in overall survival was noted between the groups (HR 0·84, 95% CI 0·64-1·12; p=0·23). Exploratory biomarker assessment suggests that patients with high pre-treatment plasma VEGFR-2 might benefit from the addition of bevacizumab (Cox interaction test p=0·029). Use of bevacizumab versus chemotherapy alone was associated with increased incidences of grade 3 or worse hypertension (154 patients [12%] vs eight patients [1%]), severe cardiac events occurring at any point during the 18-month safety reporting period (19 [1%] vs two [<0·5%]), and treatment discontinuation (bevacizumab, chemotherapy, or both; 256 [20%] vs 30 [2%]); we recorded no increase in fatal adverse events with bevacizumab (four [<0·5%] vs three [<0·5%]). INTERPRETATION Bevacizumab cannot be recommended as adjuvant treatment in unselected patients with triple-negative breast cancer. Further follow-up is needed to assess the potential effect of bevacizumab on overall survival.