POS0055 SARS-COV-2 OUTBREAK IN AUTOIMMUNE DISEASES: THE EURO-COVIMID STUDY

Background:

Coronavirus disease 2019 (COVID-19), has raised several questions in patients with immune-mediated inflammatory diseases (IMID). Whether the seroprevalence and factors associated with symptomatic COVID-19 are similar in IMID patients and in the general population is still unknown.

Objectives:

To assess the serological and clinical prevalence of COVID-19 in European IMID patients, along with the factors associated with its risk and the impacts the pandemic had on the IMID management.

Methods:

Prospective multicentre cross-sectional study among patients with five IMID (i.e. systemic lupus erythematous, Sjögren’s syndrome, rheumatoid arthritis, axial spondylarthritis or giant cell arteritis) from six tertiary-referral centers from France, Germany, Italy, Portugal, Spain and United Kingdom. Demographics, comorbidities, IMID, treatments, flares and COVID-19 details were collected. Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) serological tests were systematically performed.

Results:

Between June 7 and December 8, 2020, 3028 patients were included (median age 58 years, 73.9% females). SARS-CoV-2 antibodies were detected in 166 (5.5%) patients. Symptomatic COVID-19 was seen in 122 patients (prevalence: 4.0%, 95% CI 3.4-4.8%); 23 (24.2%) of them were hospitalized and four (3.2%) died. In multivariate logistic regression analysis, symptomatic COVID-19 was more likely to be observed in patients with higher levels of C-reactive protein (OR: 1.18; 95% CI 1.05-1.33; p = 0.006), and increased with the number of IMID flares (OR: 1.27; 95% CI 1.02-1.58; p = 0.03). Conversely, it was less likely to occur in patients treated with biological therapy (OR: 0.51; 95% CI 0.32-0.82; p = 0.006). During the pandemic, at least one self-reported disease flare was seen in 654 (21.6%) patients. Also, 519 (20.6%) patients experienced changes in their treatment, with 125 of these (24.1%) being due to COVID-19.

Conclusion:

The SARS-CoV-2 prevalence in IMID patients over the study period seems to be similar to that of the general population1. The IMID inflammatory status seems to be independently associated with the development of COVID-19.

References:

[1]Pollán M, Pérez-Gómez B, Pastor-Barriuso R, Oteo J, Hernán MA, Pérez-Olmeda M, et al. Prevalence of SARS-CoV-2 in Spain (ENE-COVID): a nationwide, population-based seroepidemiological study. Lancet Lond Engl. 2020 Aug 22;396(10250):535–44.

Disclosure of Interests:

None declared.

The value of five blood markers in differentiating mycosis fungoides and Sézary syndrome: a validation cohort

BACKGROUND

Clinical and histological diagnosis of Sézary syndrome (SS) and mycosis fungoides (MF) is challenging in clinical routine.

OBJECTIVES

We investigated five blood markers previously described for SS (T-plastin, Twist, KIR3DL2, NKp46 and Tox) in a prospective validation cohort of patients.

METHODS

We included 447 patients in this study and 107 patients were followed up for prognosis. The markers were analysed by reverse transcriptase quantitative real-time polymerase chain reaction (RT-qPCR) on peripheral blood leucocytes and CD4⁺ T cells in a cohort of consecutive patients with early MF, erythrodermic MF and SS and compared with patients presenting with benign inflammatory dermatoses (BID) and erythrodermic BID. The markers were assessed in parallel to gold standard values such as CD4/CD8 ratio, loss of CD7 and CD26 membrane expression and CD4 absolute values. Sensitivity and specificity were analysed by receiver operator characteristic curves. The prognostic value of selected markers was analysed on a subset of patients. This study was conducted in one centre.

RESULTS

We defined cut-off values for each marker. T-plastin, Twist and KIR3DL2 had the best validity. SS may be overrepresented. The combination of T-plastin and Twist was able to differentiate between erythrodermic MF or BID and SS. The additional analysis of KIR3DL2 may be useful to predict the prognosis.

CONCLUSIONS

We propose T-plastin, Twist and KIR3DL2 measured by RT-qPCR as new diagnostic markers for Sézary syndrome.

SAT0244 PROGNOSIS OF LARGE VESSEL INVOLVEMENT IN LARGE VESSEL VASCULITIS

Background:

Giant cell arteritis (GCA) and Takayasu arteritis (TAK) are the two main forms of large vessels vasculitis (LVV). Vessel inflammation leads to aneurysms, wall thickening, stenosis, and, in some cases, complete occlusion of the artery. Due to the wide variation in the course of LVV, predicting outcome is challenging. To our knowledge, data regarding prognosis factors of LVI in LVV patients and comparison of outcome of LVI in GCA and TAK are lacking. An early identification of patients with higher mortality could help to prevent deaths and vascular complications.

Objectives:

To assess prognosis factors and outcome of large vessel involvement (LVI) in large vessels vasculitis (LVV) patients.

Methods:

Retrospective multicenter study of characteristics and outcomes of 417 patients with LVI including 299 Takayasu arteritis (TAK) and 118 Giant cell arteritis (GCA-LVI) were analyzed. Logistic regression analysis assessed prognosis factors in LVV patients. Outcome of LVI among TAK and GCA-LVI patients (ischemic complications, aneurysms complications, relapses and revascularization) were assessed.

Results:

In multivariable analysis, stroke/transient ischemic attack [HR: 3.63 (1.46 - 9.04), p=0.006] was independently associated with vascular complications in LVV. The 10-years aneurysm free survival was significantly lower [67% (48 – 93) vs 89% (84 – 95), p=0.02] in GCA-LVI compare to TAK patients. The 5-years relapse free survival was significantly lower [47% (37 – 60) vs 69% (63 – 75), p<0.001,] in GCA-LVI compare to TAK patients. The 10-years revascularization free survival was significantly lower [55% (48 – 64) vs 76% (59 – 99), p<0.001] in TAK compare to GCA-LVI patients. After a median follow-up of 5 years, 16 (5.4%) TAK and 7 (5.9%) GCA-LVI patients died, mainly of aneurysm (26%) and ischemic complications (26%).

Conclusion:

This large nationwide cohort of LVI provided prognosis factors of vascular complications in LVV patients. TAK and GCA-LVI have different long-term outcome in term of aneurysm development, relapse and revascularization.

Disclosure of Interests:

None declared

SAT0587 MACHINE-LEARNING DERIVED ALGORITHMS FOR OUTCOMES PREDICTION IN RHEUMATIC DISEASES: APPLICATION TO RADIOGRAPHIC PROGRESSION IN EARLY AXIAL SPONDYLOARTHRITIS

Background:

Axial spondyloarthritis (axSpA) is a chronic rheumatic disease that encompasses various clinical presentations: inflammatory chronic back pain, peripheral manifestations and extra-articular manifestations. The current nomenclature divides axSpA in radiographic (in the presence of radiographic sacroiliitis) and non-radiographic (in the absence of radiographic sacroiliitis, with or without MRI sacroiliitis. Given that the functional burden of the disease appears to be greater in patients with radiographic forms, it seems crucial to be able to predict which patients will be more likely to develop structural damage over time. Predictive factors for radiographic progression in axSpA have been identified through use of traditional statistical models like logistic regression. However, these models present some limitations. In order to overcome these limitations and to improve the predictive performance, machine learning (ML) methods have been developed.

Objectives:

To compare ML models to traditional models to predict radiographic progression in patients with early axSpA.

Methods:

Study design: prospective French multicentric cohort study (DESIR cohort) with 5years of follow-up. Patients: all patients included in the cohort, i.e. 708 patients with inflammatory back pain for >3 months but <3 years, highly suggestive of axSpA. Data on the first 5 years of follow-up was used. Statistical analyses: radiographic progression was defined as progression either at the spine (increase of at least 1 point per 2 years of mSASSS scores) or at the sacroiliac joint (worsening of at least one grade of the mNY score between 2 visits). Traditional modelling: we first performed a bivariate analysis between our outcome (radiographic progression) and explanatory variables at baseline to select the variables to be included in our models and then built a logistic regression model (M1). Variable selection for traditional models was performed with 2 different methods: stepwise selection based on Akaike Information Criterion (stepAIC) method (M2), and the Least Absolute Shrinkage and Selection Operator (LASSO) method (M3). We also performed sensitivity analysis on all patients with manual backward method (M4) after multiple imputation of missing data. Machine learning modelling: using the “SuperLearner” package on R, we modelled radiographic progression with stepAIC, LASSO, random forest, Discrete Bayesian Additive Regression Trees Samplers (DBARTS), Generalized Additive Models (GAM), multivariate adaptive polynomial spline regression (polymars), Recursive Partitioning And Regression Trees (RPART) and Super Learner. Finally, the accuracy of traditional and ML models was compared based on their 10-foldcross-validated AUC (cv-AUC).

Results:

10-fold cv-AUC for traditional models were 0.79 and 0.78 for M2 and M3, respectively. The 3 best models in the ML algorithm were the GAM, the DBARTS and the Super Learner models, with 10-fold cv-AUC of: 0.77, 0.76 and 0.74, respectively (Table 1).

Table 1.

Comparison of 10-fold cross-validated AUC between best traditional and machine learning models.

Best modelsCross-validated AUCTraditional models M2 (step AIC method)0.79 M3 (LASSO method)0.78Machine learning approach SL Discrete Bayesian Additive Regression Trees Samplers (DBARTS)0.76 SL Generalized Additive Models (GAM)0.77 Super Learner0.74

AUC: Area Under the Curve; AIC: Akaike Information Criterion; LASSO: Least Absolute Shrinkage and Selection Operator; SL: SuperLearner. N = 295.

Conclusion:

Traditional models predicted better radiographic progression than ML models in this early axSpA population. Further ML algorithms image-based or with other artificial intelligence methods (e.g. deep learning) might perform better than traditional models in this setting.

Acknowledgments:

Thanks to the French National Society of Rheumatology and the DESIR cohort.

Disclosure of Interests:

Romain Garofoli: None declared, Matthieu resche-rigon: None declared, Maxime Dougados Grant/research support from: AbbVie, Eli Lilly, Merck, Novartis, Pfizer and UCB Pharma, Consultant of: AbbVie, Eli Lilly, Merck, Novartis, Pfizer and UCB Pharma, Speakers bureau: AbbVie, Eli Lilly, Merck, Novartis, Pfizer and UCB Pharma, Désirée van der Heijde Consultant of: AbbVie, Amgen, Astellas, AstraZeneca, BMS, Boehringer Ingelheim, Celgene, Cyxone, Daiichi, Eisai, Eli-Lilly, Galapagos, Gilead Sciences, Inc., Glaxo-Smith-Kline, Janssen, Merck, Novartis, Pfizer, Regeneron, Roche, Sanofi, Takeda, UCB Pharma; Director of Imaging Rheumatology BV, Christian Roux: None declared, Anna Moltó Grant/research support from: Pfizer, UCB, Consultant of: Abbvie, BMS, MSD, Novartis, Pfizer, UCB

AB0508 EFFICACY AND SAFETY OF TNF-Α ANTAGONISTS AND TOCILIZUMAB IN TAKAYASU ARTERITIS: MULTICENTER EUROPEAN RETROSPECTIVE STUDY OF 203 PATIENTS

Background:

Takayasu arteritis (TA) is a chronic inflammatory large-vessel vasculitis, predominantly affecting the aorta and its main branches

Objectives:

To assess safety and efficacy of biologics (i.e. TNF-α antagonists and tocilizumab) in patients with Takayasu arteritis (TA).

Methods:

We conducted a retrospective multicenter study in referral centers from Europe and several countries in the world about biological-targeted therapies in Takayasu arteritis during the period from January 2017 to September 2019.

Results:

Retrospective multicenter study of characteristics and outcome of 49 TA patients [80% of females; median age 42 [20-55] years] treated by TNF-α antagonists (80%) or tocilizumab (20%)] and fulfilling ACR and/or Ishikawa criteria. Factors associated with complete response were assessed. Eighty-eight percent of TA patients were inadequately controlled with, or intolerant to, conventional immunosuppressive therapy [median number of 3 (1-5)]. Overall response (i.e. complete and partial) to biological-targeted treatments at 6 and 12 months was of 75% and 83%, respectively. There were a significantly lower C-reactive protein levels at initiation of biological-targeted treatments [22 [10-46] mg/l vs 58 [26-76] mg/l, (p=0.006)] and a trend toward lower immunosuppressants drugs used prior biologics (p=0.054) in responders (i.e. complete and/or partial responders) relative to non-responders to biological-targeted treatments. C-reactive protein levels and daily prednisone dosage significantly decreased after 12 months of biological-targeted treatments [30 vs 6 mg/l, p<0.05 and 15 vs 7.5 mg, p<0.05, at baseline and at 12 months, respectively]. The 3-year relapse free survival was of 90.9% (83.5-99) over biologic treatment period compared to 58.7% (43.3-79.7) (p=0.0025) with DMARDs. No difference was found relative to efficacy between TNF-α antagonists and tocilizumab. After a median follow-up of 24 [2-95] months, 21% of adverse effects occurred, with biological-targeted treatments discontinuation in 6.6% of cases.

Conclusion:

This nationwide study shows high efficacy of biological-targeted treatments in refractory TA patients with an acceptable safety profile.

Acknowledgments:

NO

Disclosure of Interests:

Arsene Mekinian: None declared, Mathieu Vautier: None declared, Matthieu resche-rigon: None declared, Lorenzo Dagna Grant/research support from: Abbvie, BMS, Celgene, Janssen, MSD, Mundipharma Pharmaceuticals, Novartis, Pfizer, Roche, SG, SOBI, Consultant of: Abbvie, Amgen, Biogen, BMS, Celltrion, Novartis, Pfizer, Roche, SG, and SOBI, Savino Sciascia: None declared, olivier.espitia@chu-nantes.fr olivier.espitia@chu-nantes.fr: None declared, Patrice cacoub: None declared, olivier fain: None declared, david Saadoun: None declared

OP0146 LONG TERM OUTCOME AND PROGNOSIS FACTORS OF ISOLATED AORTITIS

Background:

Aortitis is a group of disorders characterized by the inflammation of the aorta. The most common causes of aortitis are the large-vessel vasculitis i.e. giant cell arteritis (GCA) and Takayasu arteritis (TA). However, aortitis may be isolated. Because of the wide variation in the course of aortitis, predicting outcome is challenging. The optimal management strategy of isolated aortitis (IA) is still unclear as IA is poorly defined, with data consisting of small retrospective and case control studies.

Objectives:

To assess the long-term outcome and prognosis factors for vascular complications in patients with isolated aortitis.

Methods:

Retrospective multicenter study of 353 patients with non-infectious aortitis including 136 giant cell arteritis (GCA), 96 Takayasu arteritis (TA) and 73 isolated aortitis (IA). Factors associated with event-free survival, vascular event-free survival and revascularization-free survival were assessed. Risk factors for vascular complications were identified in multivariate analysis.

Results:

After a median follow up of 52 months, vascular complications were observed in 32.3 %, revascularization in 30 % and death in 7.6%. The 5-year cumulative incidence of vascular complications was 58% (41; 71), 20% (13; 29), and 19 % (11; 28) in IA, GCA and TA, respectively. In multivariate analysis, IA [HR, 1.85 (1.19 to 2.88), p=0.017] and male gender [1.77 (1.26 to 2.49), p<0.0001] were independently associated with vascular events. The 5-year surgery-free survival was 45% (31; 65), 71% (62; 81) and 76% (68; 86) in IA, TA and GCA, respectively.

Conclusion:

IA has a worse vascular prognosis than GCA and TA. Sixty percent of IA patients will experience a vascular complication within 5 years from diagnosis.

Disclosure of Interests:

None declared

Validité de la FRENCH, l’échelle de tri des urgences hospitalières élaborée par la Société française de médecine d’urgence

Introduction : La Société française de médecine d’urgence a créé en 2016 une échelle de tri pour les infirmier( ière)s d’accueil dans une structure d’urgences. Cette échelle appelée FRENCH (FRench Emergency Nurses Classification in Hospital) classe les tris de 5 à 1 (du moins urgent au plus urgent) en fonction du pronostic et de la complexité/ sévérité des motifs de recours aux soins. Le tri 3, groupe hétérogène dans les échelles de tri internationales, a été subdivisé en deux niveaux pour prioriser les patients qui ont une comorbidité en rapport avec le motif de recours aux soins ou qui sont adressés par un médecin et qui sont prioritaires (3A) par rapport aux autres patients (3B).

Objectif : Évaluer la pertinence de la FRENCH.

Méthode : Nous avons analysé les données démographiques, les paramètres vitaux, les examens complémentaires prescrits et les durées de prise en charge de tous les patients accueillis dans un hôpital universitaire sur une période de neuf mois consécutifs. Le critère de jugement était l’existence d’une relation croissante entre le niveau de complexité/ sévérité des patients — reflétée par le taux d’hospitalisation et la prescription d’examens complémentaires—et le niveau de priorité du tri.

Résultats : L’étude a inclus 27 598 patients. La répartition des patients par niveaux de tri était : 0,4 (tri 1), 6,7 (tri 2), 13,3 (tri 3A), 29,4 (tri 3B), 43,1 (tri 4) et 7,1 % (tri 5). Le taux d’hospitalisation était croissant quand l’ordre de priorité augmentait. La relation entre les niveaux de tri et le taux d’hospitalisation mesurée par l’aire sous la courbe (0,83 : intervalle de confiance à 95 % : [0,82–0,83]) était bonne. La prescription des examens complémentaires était croissante quand l’ordre de priorité augmentait. La corrélation entre les niveaux de tri et un indice global d’examens complémentaires prescrits était modérée (K = 0,51).

Conclusion : Cette étude monocentrique valide la pertinence de l’échelle de tri FRENCH à six niveaux par sa bonne capacité à classer les patients selon leur complexité/sévérité. De nouvelles évaluations sont nécessaires dans d’autres structures d’urgences pour confirmer sa performance et favoriser son évolution.

Erratum to: 36th International Symposium on Intensive Care and Emergency Medicine: Brussels, Belgium. 15-18 March 2016

[This corrects the article DOI: 10.1186/s13054-016-1208-6.].

Usefulness of daily surveillance blood cultures in allogeneic hematopoietic stem cell transplant recipients on steroids: a 1-year prospective study

BACKGROUND

Bloodstream infections (BSI) are frequent and potentially severe complications in allogeneic hematopoietic stem cell transplant (AHSCT) recipients. In patients on steroids, surveillance blood cultures (SBCs) are routinely performed to detect asymptomatic BSI but their usefulness remains controversial.

METHODS

We performed a 1-year, observational, prospective, single-center study to assess the utility of daily SBCs in AHSCT recipients on steroids and a case-control study to identify risk factors associated with positive SBCs. All blood cultures (BCs) obtained from adults hospitalized in the HSCT unit were prospectively studied throughout 1 year. Characteristics, treatments, and outcome of patients were retrieved from medical charts.

RESULTS

A total of 3594 BCs were obtained in 177 patients, including 1450 SBCs in 82 AHSCT recipients on steroids. In 33 patients, 103 SBCs (7%) were positive. Low-virulence bacteria were identified in 74% of episodes. When analyzing first episode of positive SBCs (28 patients), 6 (21%) true BSI were identified.

CONCLUSIONS

Patients with positive SBCs were receiving antibiotic treatment less frequently at the time of SBCs (P < 0.001) and had more frequently BCs obtained through central venous access (P < 0.04) when compared to patients with negative SBCs. Daily SBCs in AHSCT recipients on steroids only rarely identify BSI and clear benefit for patients could not be demonstrated.

Angle of fetal head progression measured using transperineal ultrasound as a predictive factor of vacuum extraction failure

OBJECTIVE

To evaluate the predictive value of angle of progression (AoP) of the fetal head for a failed vacuum delivery.

METHODS

This was a prospective observational study that included women with a singleton pregnancy of ≥ 37 weeks' gestation, in cephalic presentation requiring vacuum extraction. Transperineal ultrasound was performed immediately before vacuum extraction, although AoP was measured on stored images after delivery. Vacuum extraction was defined as failed when the duration of extraction exceeded 20 min or the vacuum cup detached more than three times. We compared the demographic and ultrasound data of failed vacuum deliveries with those that were successful. The predictive value of AoP for failure of vacuum delivery was calculated.

RESULTS

AoP was measured in 235 women. Vacuum extractions failed in 30 (12.8%) women (29/184 nulliparous and 1/51 parous) and resulted in 28 vaginal deliveries by forceps and two Cesarean deliveries. Median AoP was significantly lower in the vacuum failure group compared with those with successful vacuum delivery (136.6° (interquartile range (IQR), 129.8-144.1°) vs 145.9° (IQR, 135.0-158.4°); P < 0.01). As all but one failed vacuum extraction occurred among nulliparous women, the predictive value of AoP was calculated in this subgroup of women. The area under the receiver-operating characteristics curve for prediction of vacuum extraction failure was 0.67 (95% CI, 0.57-0.77) and the optimal AoP cut-off was 145.5°. Above this value, the rate of vacuum extraction failure fell below 5%.

CONCLUSION

AoP is a predictive factor of failed vacuum extraction, especially among nulliparous women whose risk of failure is high. AoP measurement may help in choosing between forceps and vacuum extraction. Copyright © 2015 ISUOG. Published by John Wiley & Sons Ltd.

Investigating covariate-by-centre interaction in survival data

In survival analysis, assessing the existence of potential centre effects on the baseline hazard or on the effect of fixed covariates on the baseline hazard, such as treatment-by-centre interaction, is a frequent clinical concern in multicentre studies. Survival models with random effects on the baseline hazard and/or on the effect of the covariates of interest have been largely applied, for instance, to investigate potential centre effects. We aimed to develop a procedure to routinely test for multiple random effects in survival analyses. We propose a statistic and a permutation approach to test whether all or a subset of components of the variance-covariance matrix of random effects are non-zero in a mixed-effects Cox model framework. Performances of the proposed permutation tests are examined under different null hypotheses corresponding to the different components of the variance-covariance matrix, i.e ., to the different random effects considered on the baseline hazard and/or on the covariates effects. Several alternative hypotheses are evaluated using simulations. The results indicate that the permutation tests have valid type I error rates under the null and achieve satisfactory power under all alternatives. The procedure is applied to two European cohorts of haematological stem cell transplants in acute leukaemia to investigate the heterogeneity across centres in leukaemia-free survival and the potential heterogeneity in prognostic factors effects across centres.

Molecular determinants of the recognition of ulipristal acetate by oxo-steroid receptors

The human progesterone receptor (PR) plays a key role in reproductive function in women. PR antagonists have numerous applications in female health care including regular and emergency contraception, and treatment of hormone-related pathological conditions such as breast cancer, endometriosis, and leiomyoma. The main factor limiting their long-term administration is the fact that they cross-bind to other oxo-steroid receptors. Ulipristal acetate (UPA), a highly potent PR antagonist, has recently come onto the market and is much more selective for PR than the other oxo-steroid receptors (androgen, AR, glucocorticoid, GR, and mineralocorticoid, MR receptors) and, remarkably, it displays lower GR-inactivating potency than RU486. We adopted a structural approach to characterizing the binding of UPA to the oxo-steroid receptors at the molecular level. We solved the X-ray crystal structure of the ligand-binding domain (LBD) of the human PR complexed with UPA and a peptide from the transcriptional corepressor SMRT. We used the X-ray crystal structure of the GR in its antagonist conformation to dock UPA within its ligand-binding cavity. Finally, we generated three-dimensional models of the LBD of androgen and mineralocorticoid receptors (AR and MR) in an antagonist conformation and docked UPA within them. Comparing the structures revealed that the network of stabilizing contacts between the UPA C11 aryl group and the LBD is responsible for its high PR antagonist potency. It also showed that it is the inability of UPA to contact Gln642 in GR that explains why it has lower potency in inactivating GR than RU486. Finally, we found that the binding pockets of AR and MR are too small to accommodate UPA, and allowed us to propose that the extremely low sensitivity of MR to UPA is due to inappropriate interactions with the C11 substituent. All these findings open new avenues for designing new PR antagonist compounds displaying greater selectivity.

Trends in death attributed to myocardial infarction, heart failure and pulmonary embolism in Europe and Canada over the last decade

BACKGROUND

Worldwide, cardiovascular diseases and cancer account for ∼40% of deaths. Certain reports have shown a progressive decrease in mortality. Our main objective was to assess mortality trends related to myocardial infarction (MI), heart failure (HF) and pulmonary embolism (PE).

METHODS

MI, HF and PE were studied as cause of death based on the analysis of death certificates in Canada (C), England and Wales (E), France (F) and Sweden (S). We also used a multiple cause approach. Age-standardized death rates (SDR) were calculated.

RESULTS

The SDR for MI, HF or PE as the underlying cause of death, all decreased during the last decade. The decrease in SDR secondary to MI exceeded that for HF or PE. Concerning multiple cause of death, a greater decrease was also found for MI, compared with HF or PE.

CONCLUSIONS

We confirm the beneficial trends in SDR with MI, HF or PE both as underlying or multiple causes in the studied countries. For HF and PE, multiple cause approach seems more accurate to describe the burden of these two pathologies. Our study also suggests that more efforts should be dedicated to HF and PE in order to achieve similar trends than in MI.

Permutation tests for centre effect on survival endpoints with application in an acute myeloid leukaemia multicentre study

When analysing multicentre data, it may be of interest to test whether the distribution of the endpoint varies among centres. In a mixed-effect model, testing for such a centre effect consists in testing to zero a random centre effect variance component. It has been shown that the usual asymptotic χ(2) distribution of the likelihood ratio and score statistics under the null does not necessarily hold. In the case of censored data, mixed-effects Cox models have been used to account for random effects, but few works have concentrated on testing to zero the variance component of the random effects. We propose a permutation test, using random permutation of the cluster indices, to test for a centre effect in multilevel censored data. Results from a simulation study indicate that the permutation tests have correct type I error rates, contrary to standard likelihood ratio tests, and are more powerful. The proposed tests are illustrated using data of a multicentre clinical trial of induction therapy in acute myeloid leukaemia patients.

Diagnostic accuracy of urinary dipstick to exclude catheter-associated urinary tract infection in ICU patients: a reappraisal

OBJECTIVES

We wanted to assess the diagnostic accuracy of urinary dipstick testing in excluding catheter-associated urinary tract infection (CAUTI) in intensive care unit (ICU) patients with fever or hypothermia.

METHODS

This was a prospective observational cohort study in a medical-surgical ICU. Patients with new-onset fever >38.3 °C or hypothermia <36 °C at least 48 h after urinary catheter insertion were included over a 2-year period. At each episode, a urinary dipstick test and a urine culture were performed as the criterion standard. Extensive microbiological investigations for extra-urinary infections were performed also. The performances of various urinary dipstick result combinations in ruling out CAUTI were compared based on the likelihood ratios (LR+ and LR-).

RESULTS

Symptomatic CAUTI was diagnosed in 31 (24.4 %) of the 127 included patients (195 episodes of fever or hypothermia). LR+ was best for combined leukocyte esterase-positive and nitrite-positive dipstick results (overall population: 14.91; 95 % confidence interval [95 % CI], 5.53-40.19; patients without urinary symptoms: 15.63; 95 % CI, 5.76-42.39). LR- was best for either leukocyte esterase-positive or nitrite-positive dipstick results (overall population: 0.41; 95 % CI, 0.57-0.65; patients without urinary symptoms, 0.36; 95 % CI, 0.21-0.60).

CONCLUSIONS

Urinary dipstick testing at the bedside does not help to rule out symptomatic CAUTI in medical or surgical ICU patients with fever or hypothermia.

Number of examined lymph nodes and nodal status assessment in pancreaticoduodenectomy for pancreatic adenocarcinoma

BACKGROUND

The accuracy of the assessment of the nodal status in resected cephalic pancreatic adenocarcinoma (PA) depends on the number of examined lymph nodes (NELN). This study assesses the impact of the NELN on N staging and survival and propose a minimal number of examined lymph nodes (MNELN) ensuring reliability of the pN status determination.

METHODS

188 consecutive patients treated by pancreaticoduodenectomy (PD) for PA. Correlations between NELN and survivals of pN0 and pN1 groups and with the rate of pN1 patients were studied. A probability model based on the binomial law was built to estimate the MNELN able to detect pN1 patients with a sensitivity ≥ 95%.

RESULTS

Overall and disease free 5-year survivals were 27.2% and 24.6% respectively. 135 patients (71.8%) were staged pN1. The median NELN was 17 (range 0-68). Overall and disease free survivals of pN1 patients were not related to NELN. The influence of NELN on survival in pN0 patients due to stage migration did not reach significance. The probability model showed that a MNELN of 16 nodes was required to detect pN1 patients with a sensitivity of 95%.

CONCLUSION

A MNELN of 16 is required to assess pN status and should be considered as a quality criterion in future studies and trials on PD for PA.