PMC:6964229 / 3092-10561 JSONTXT

{"project":"0_colil","denotations":[{"id":"31203374-18374749-6896","span":{"begin":1267,"end":1268},"obj":"18374749"},{"id":"31203374-16427840-6897","span":{"begin":1621,"end":1622},"obj":"16427840"},{"id":"31203374-16103039-6898","span":{"begin":3703,"end":3704},"obj":"16103039"}],"text":"METHODS\n\nPatients\nAll adult patients (≥18 years of age) who underwent surgical intervention (n = 528) for MR due to degenerative disease between January 2004 and December 2015 at our institution were eligible for inclusion. We excluded patients with a history of previous cardiac surgery (n = 18) and patients in whom MV replacement was performed (n = 9). Of the remaining 501 patients, 17 additional patients were excluded for the following reasons: early mortality (n = 6), non-use of annuloplasty ring (n = 4) or missing predischarge echocardiogram (n = 7). The final study cohort compromised 484 patients who all underwent successful MV repair with a complete annuloplasty ring.\nMeasurements of MV gradients were acquired from continuous wave Doppler acquisitions of the diastolic inflow of the MV. Mean MV gradient was calculated as the average of 3 and 5 cycles or patients in sinus rhythm and atrial fibrillation, respectively. All acquisitions were performed according to a standardized protocol of our echo lab to ensure data reproducibility. All measurements were performed by experienced echocardiographists.\n\nStudy end points\nAll end points were defined according to the Guidelines for reporting mortality and morbidity after cardiac valve interventions [6]. Primary end points included all-cause mortality and freedom from MV reintervention. Secondary end points included freedom from atrial fibrillation. We excluded other types of atrial tachycardias (e.g. atrial flutter or incisional atrial tachycardia) because of aetiological differences associated with the development postrepair atrial tachycardias [7]. Only patients in preoperative sinus rhythm (n = 312) were included in the latter analysis.\n\nSurgical technique and perioperative care\nWhen indicated, anterior MV leaflet prolapse was addressed with chordal replacement. Posterior MV leaflet prolapse was addressed with a combination of leaflet resection and chordal replacement techniques. The decision on which technique to use was based on the extent of leaflet prolapse and excessive tissue in height and/or width. Earlier in this series, annular plication was more frequently used to help restore the continuity of the posterior MV leaflet. Later, leaflet sliding techniques were employed. Commissural prolapse was primarily addressed with papillary muscle head repositioning.\nAll patients included in the current study underwent full annuloplasty ring implantation. Ring sizing was standardly based on the surface area of the anterior MV leaflet and was not influenced by the type of anterior and/or posterior MV leaflet repair technique used. No over- or downsizing was performed in any of the patients.\nIntraoperative transoesophageal echocardiography was performed by an experienced cardiologist to document the intraoperative result of MV repair. Additionally, predischarge transthoracic echocardiography (performed on postoperative day 4–7) was performed to exclude any significant residual MR. Systolic anterior motion was observed in 2 patients and treated conservatively. Oral anticoagulation was initiated for a period of 3 months with a target Internationalized Normalized Ratio of 2.0–3.0. In presence of other indications, the target Internationalized Normalized Ratio range and treatment duration were adjusted accordingly.\n\nFollow-up\nPreoperative, intraoperative and postoperative data were retrospectively collected from our computerized database. Follow-up survival, clinical and echocardiographic data were collected through regular clinical visits at our institution or affiliated clinics and hospitals, and through questionnaires to patients. Patient follow-up was based on the available recommendations [8]. Records pertaining to reported office visits, echocardiography and rhythm information, operations, cardioversions, catheter ablation or hospitalizations were obtained and analysed. Rhythm follow-up was based on electrocardiograms obtained during regular follow-up. Additional rhythm monitoring (e.g. 24-h Holter monitoring) was performed on clinical grounds (e.g. complaints of heart palpitations) and the indication for this was left at the discretion of the attending cardiologist. A total of 1126 (mean 2.3/patient, range 0–8) follow-up echocardiograms were available for analysis.\nThe study was approved by our Institutional Ethics Committee and patient consent was obtained to allow collection and analysis of any relevant clinical and echocardiographic data. Follow-up was completed in February 2017.\n\nStatistical analysis\nContinuous data are presented as means ± standard deviation for normally distributed data or medians and interquartile range (IQR) when not normally distributed. Categorical data are presented as counts and percentages. The distribution of variables was evaluated using the Kolmogorov–Smirnov test.\nA univariable and multivariable linear regression model, with postrepair MV gradient as a dependent continuous variable, was built to explore the factors associated with postrepair MV gradient. Variable inclusion was based on the previously established effect on MV gradient. To assess for the presence of violations in model assumptions, residuals were plotted versus fitted values and investigated graphically. Univariable and multivariable Cox proportional hazards regression analysis was used to explore risk factors for the occurrence of time-to-event outcomes. Based on known clinical validity and taking into account the ratio of events to risk factors, the following factors were included in respective multivariate models: ‘for mortality’: age, gender, chronic pulmonary disease, renal impairment, left ventricular function, symptomatic MR, atrial fibrillation and postrepair MV gradient; ‘for reintervention’: anterior MV leaflet repair and postrepair MV gradient; and ‘for atrial fibrillation’: age, tricuspid valve repair, preoperative left atrial diameter and postrepair MV gradient. Cubic third-degree splines were used to adjust for possible non-linear association of mean MV gradient with time-to-event outcome in the proportional hazards model. The Wald test was used to test for the presence of non-linear effect. In addition to P-value-based hypothesis testing, non-linearity of mean MV gradient was assessed graphically by plotting the fitted splines for mean MV gradient, together with 95% confidence intervals (CIs) on the log-hazard scale versus mean MV gradient.\nTo study the evolution of mean MV gradients during the follow-up period, a mixed-model based linear regression analysis of MV gradient on follow-up time which accounts for repeated within-patient observation was carried out. To account for potential heteroscedasticity or non-normality, bootstrap-based standard errors were investigated. In addition, a post hoc exploration of model residuals was explored to investigate these same model assumptions as well as the linearity assumption of MV gradient decent versus follow-up time. No evidence of serious model deviations was found and hence the regular standard errors and hypothesis test are reported.\nAll tests were 2-tailed and a P-value of \u003c0.05 was considered statistically significant. Statistical analysis was performed using IBM Statistics for Windows, version 23.0 (IBM Corp. Released 2015. IBM SPSS Statistics for Windows, Version 23.0. Armonk, NY, USA: IBM Corp.) and R, version 3.5.1 using the Survival package, version 2.43–3 (R Foundation for Statistical Computing, Vienna, Austria)."}

{"project":"2_test","denotations":[{"id":"31203374-18374749-28905062","span":{"begin":1267,"end":1268},"obj":"18374749"},{"id":"31203374-16427840-28905063","span":{"begin":1621,"end":1622},"obj":"16427840"},{"id":"31203374-16103039-28905064","span":{"begin":3703,"end":3704},"obj":"16103039"}],"text":"METHODS\n\nPatients\nAll adult patients (≥18 years of age) who underwent surgical intervention (n = 528) for MR due to degenerative disease between January 2004 and December 2015 at our institution were eligible for inclusion. We excluded patients with a history of previous cardiac surgery (n = 18) and patients in whom MV replacement was performed (n = 9). Of the remaining 501 patients, 17 additional patients were excluded for the following reasons: early mortality (n = 6), non-use of annuloplasty ring (n = 4) or missing predischarge echocardiogram (n = 7). The final study cohort compromised 484 patients who all underwent successful MV repair with a complete annuloplasty ring.\nMeasurements of MV gradients were acquired from continuous wave Doppler acquisitions of the diastolic inflow of the MV. Mean MV gradient was calculated as the average of 3 and 5 cycles or patients in sinus rhythm and atrial fibrillation, respectively. All acquisitions were performed according to a standardized protocol of our echo lab to ensure data reproducibility. All measurements were performed by experienced echocardiographists.\n\nStudy end points\nAll end points were defined according to the Guidelines for reporting mortality and morbidity after cardiac valve interventions [6]. Primary end points included all-cause mortality and freedom from MV reintervention. Secondary end points included freedom from atrial fibrillation. We excluded other types of atrial tachycardias (e.g. atrial flutter or incisional atrial tachycardia) because of aetiological differences associated with the development postrepair atrial tachycardias [7]. Only patients in preoperative sinus rhythm (n = 312) were included in the latter analysis.\n\nSurgical technique and perioperative care\nWhen indicated, anterior MV leaflet prolapse was addressed with chordal replacement. Posterior MV leaflet prolapse was addressed with a combination of leaflet resection and chordal replacement techniques. The decision on which technique to use was based on the extent of leaflet prolapse and excessive tissue in height and/or width. Earlier in this series, annular plication was more frequently used to help restore the continuity of the posterior MV leaflet. Later, leaflet sliding techniques were employed. Commissural prolapse was primarily addressed with papillary muscle head repositioning.\nAll patients included in the current study underwent full annuloplasty ring implantation. Ring sizing was standardly based on the surface area of the anterior MV leaflet and was not influenced by the type of anterior and/or posterior MV leaflet repair technique used. No over- or downsizing was performed in any of the patients.\nIntraoperative transoesophageal echocardiography was performed by an experienced cardiologist to document the intraoperative result of MV repair. Additionally, predischarge transthoracic echocardiography (performed on postoperative day 4–7) was performed to exclude any significant residual MR. Systolic anterior motion was observed in 2 patients and treated conservatively. Oral anticoagulation was initiated for a period of 3 months with a target Internationalized Normalized Ratio of 2.0–3.0. In presence of other indications, the target Internationalized Normalized Ratio range and treatment duration were adjusted accordingly.\n\nFollow-up\nPreoperative, intraoperative and postoperative data were retrospectively collected from our computerized database. Follow-up survival, clinical and echocardiographic data were collected through regular clinical visits at our institution or affiliated clinics and hospitals, and through questionnaires to patients. Patient follow-up was based on the available recommendations [8]. Records pertaining to reported office visits, echocardiography and rhythm information, operations, cardioversions, catheter ablation or hospitalizations were obtained and analysed. Rhythm follow-up was based on electrocardiograms obtained during regular follow-up. Additional rhythm monitoring (e.g. 24-h Holter monitoring) was performed on clinical grounds (e.g. complaints of heart palpitations) and the indication for this was left at the discretion of the attending cardiologist. A total of 1126 (mean 2.3/patient, range 0–8) follow-up echocardiograms were available for analysis.\nThe study was approved by our Institutional Ethics Committee and patient consent was obtained to allow collection and analysis of any relevant clinical and echocardiographic data. Follow-up was completed in February 2017.\n\nStatistical analysis\nContinuous data are presented as means ± standard deviation for normally distributed data or medians and interquartile range (IQR) when not normally distributed. Categorical data are presented as counts and percentages. The distribution of variables was evaluated using the Kolmogorov–Smirnov test.\nA univariable and multivariable linear regression model, with postrepair MV gradient as a dependent continuous variable, was built to explore the factors associated with postrepair MV gradient. Variable inclusion was based on the previously established effect on MV gradient. To assess for the presence of violations in model assumptions, residuals were plotted versus fitted values and investigated graphically. Univariable and multivariable Cox proportional hazards regression analysis was used to explore risk factors for the occurrence of time-to-event outcomes. Based on known clinical validity and taking into account the ratio of events to risk factors, the following factors were included in respective multivariate models: ‘for mortality’: age, gender, chronic pulmonary disease, renal impairment, left ventricular function, symptomatic MR, atrial fibrillation and postrepair MV gradient; ‘for reintervention’: anterior MV leaflet repair and postrepair MV gradient; and ‘for atrial fibrillation’: age, tricuspid valve repair, preoperative left atrial diameter and postrepair MV gradient. Cubic third-degree splines were used to adjust for possible non-linear association of mean MV gradient with time-to-event outcome in the proportional hazards model. The Wald test was used to test for the presence of non-linear effect. In addition to P-value-based hypothesis testing, non-linearity of mean MV gradient was assessed graphically by plotting the fitted splines for mean MV gradient, together with 95% confidence intervals (CIs) on the log-hazard scale versus mean MV gradient.\nTo study the evolution of mean MV gradients during the follow-up period, a mixed-model based linear regression analysis of MV gradient on follow-up time which accounts for repeated within-patient observation was carried out. To account for potential heteroscedasticity or non-normality, bootstrap-based standard errors were investigated. In addition, a post hoc exploration of model residuals was explored to investigate these same model assumptions as well as the linearity assumption of MV gradient decent versus follow-up time. No evidence of serious model deviations was found and hence the regular standard errors and hypothesis test are reported.\nAll tests were 2-tailed and a P-value of \u003c0.05 was considered statistically significant. Statistical analysis was performed using IBM Statistics for Windows, version 23.0 (IBM Corp. Released 2015. IBM SPSS Statistics for Windows, Version 23.0. Armonk, NY, USA: IBM Corp.) and R, version 3.5.1 using the Survival package, version 2.43–3 (R Foundation for Statistical Computing, Vienna, Austria)."}

{"project":"MyTest","denotations":[{"id":"31203374-18374749-28905062","span":{"begin":1267,"end":1268},"obj":"18374749"},{"id":"31203374-16427840-28905063","span":{"begin":1621,"end":1622},"obj":"16427840"},{"id":"31203374-16103039-28905064","span":{"begin":3703,"end":3704},"obj":"16103039"}],"namespaces":[{"prefix":"_base","uri":"https://www.uniprot.org/uniprot/testbase"},{"prefix":"UniProtKB","uri":"https://www.uniprot.org/uniprot/"},{"prefix":"uniprot","uri":"https://www.uniprot.org/uniprotkb/"}],"text":"METHODS\n\nPatients\nAll adult patients (≥18 years of age) who underwent surgical intervention (n = 528) for MR due to degenerative disease between January 2004 and December 2015 at our institution were eligible for inclusion. We excluded patients with a history of previous cardiac surgery (n = 18) and patients in whom MV replacement was performed (n = 9). Of the remaining 501 patients, 17 additional patients were excluded for the following reasons: early mortality (n = 6), non-use of annuloplasty ring (n = 4) or missing predischarge echocardiogram (n = 7). The final study cohort compromised 484 patients who all underwent successful MV repair with a complete annuloplasty ring.\nMeasurements of MV gradients were acquired from continuous wave Doppler acquisitions of the diastolic inflow of the MV. Mean MV gradient was calculated as the average of 3 and 5 cycles or patients in sinus rhythm and atrial fibrillation, respectively. All acquisitions were performed according to a standardized protocol of our echo lab to ensure data reproducibility. All measurements were performed by experienced echocardiographists.\n\nStudy end points\nAll end points were defined according to the Guidelines for reporting mortality and morbidity after cardiac valve interventions [6]. Primary end points included all-cause mortality and freedom from MV reintervention. Secondary end points included freedom from atrial fibrillation. We excluded other types of atrial tachycardias (e.g. atrial flutter or incisional atrial tachycardia) because of aetiological differences associated with the development postrepair atrial tachycardias [7]. Only patients in preoperative sinus rhythm (n = 312) were included in the latter analysis.\n\nSurgical technique and perioperative care\nWhen indicated, anterior MV leaflet prolapse was addressed with chordal replacement. Posterior MV leaflet prolapse was addressed with a combination of leaflet resection and chordal replacement techniques. The decision on which technique to use was based on the extent of leaflet prolapse and excessive tissue in height and/or width. Earlier in this series, annular plication was more frequently used to help restore the continuity of the posterior MV leaflet. Later, leaflet sliding techniques were employed. Commissural prolapse was primarily addressed with papillary muscle head repositioning.\nAll patients included in the current study underwent full annuloplasty ring implantation. Ring sizing was standardly based on the surface area of the anterior MV leaflet and was not influenced by the type of anterior and/or posterior MV leaflet repair technique used. No over- or downsizing was performed in any of the patients.\nIntraoperative transoesophageal echocardiography was performed by an experienced cardiologist to document the intraoperative result of MV repair. Additionally, predischarge transthoracic echocardiography (performed on postoperative day 4–7) was performed to exclude any significant residual MR. Systolic anterior motion was observed in 2 patients and treated conservatively. Oral anticoagulation was initiated for a period of 3 months with a target Internationalized Normalized Ratio of 2.0–3.0. In presence of other indications, the target Internationalized Normalized Ratio range and treatment duration were adjusted accordingly.\n\nFollow-up\nPreoperative, intraoperative and postoperative data were retrospectively collected from our computerized database. Follow-up survival, clinical and echocardiographic data were collected through regular clinical visits at our institution or affiliated clinics and hospitals, and through questionnaires to patients. Patient follow-up was based on the available recommendations [8]. Records pertaining to reported office visits, echocardiography and rhythm information, operations, cardioversions, catheter ablation or hospitalizations were obtained and analysed. Rhythm follow-up was based on electrocardiograms obtained during regular follow-up. Additional rhythm monitoring (e.g. 24-h Holter monitoring) was performed on clinical grounds (e.g. complaints of heart palpitations) and the indication for this was left at the discretion of the attending cardiologist. A total of 1126 (mean 2.3/patient, range 0–8) follow-up echocardiograms were available for analysis.\nThe study was approved by our Institutional Ethics Committee and patient consent was obtained to allow collection and analysis of any relevant clinical and echocardiographic data. Follow-up was completed in February 2017.\n\nStatistical analysis\nContinuous data are presented as means ± standard deviation for normally distributed data or medians and interquartile range (IQR) when not normally distributed. Categorical data are presented as counts and percentages. The distribution of variables was evaluated using the Kolmogorov–Smirnov test.\nA univariable and multivariable linear regression model, with postrepair MV gradient as a dependent continuous variable, was built to explore the factors associated with postrepair MV gradient. Variable inclusion was based on the previously established effect on MV gradient. To assess for the presence of violations in model assumptions, residuals were plotted versus fitted values and investigated graphically. Univariable and multivariable Cox proportional hazards regression analysis was used to explore risk factors for the occurrence of time-to-event outcomes. Based on known clinical validity and taking into account the ratio of events to risk factors, the following factors were included in respective multivariate models: ‘for mortality’: age, gender, chronic pulmonary disease, renal impairment, left ventricular function, symptomatic MR, atrial fibrillation and postrepair MV gradient; ‘for reintervention’: anterior MV leaflet repair and postrepair MV gradient; and ‘for atrial fibrillation’: age, tricuspid valve repair, preoperative left atrial diameter and postrepair MV gradient. Cubic third-degree splines were used to adjust for possible non-linear association of mean MV gradient with time-to-event outcome in the proportional hazards model. The Wald test was used to test for the presence of non-linear effect. In addition to P-value-based hypothesis testing, non-linearity of mean MV gradient was assessed graphically by plotting the fitted splines for mean MV gradient, together with 95% confidence intervals (CIs) on the log-hazard scale versus mean MV gradient.\nTo study the evolution of mean MV gradients during the follow-up period, a mixed-model based linear regression analysis of MV gradient on follow-up time which accounts for repeated within-patient observation was carried out. To account for potential heteroscedasticity or non-normality, bootstrap-based standard errors were investigated. In addition, a post hoc exploration of model residuals was explored to investigate these same model assumptions as well as the linearity assumption of MV gradient decent versus follow-up time. No evidence of serious model deviations was found and hence the regular standard errors and hypothesis test are reported.\nAll tests were 2-tailed and a P-value of \u003c0.05 was considered statistically significant. Statistical analysis was performed using IBM Statistics for Windows, version 23.0 (IBM Corp. Released 2015. IBM SPSS Statistics for Windows, Version 23.0. Armonk, NY, USA: IBM Corp.) and R, version 3.5.1 using the Survival package, version 2.43–3 (R Foundation for Statistical Computing, Vienna, Austria)."}

{"project":"testtesttest","denotations":[{"id":"T27","span":{"begin":318,"end":320},"obj":"Body_part"},{"id":"T28","span":{"begin":638,"end":640},"obj":"Body_part"},{"id":"T29","span":{"begin":699,"end":701},"obj":"Body_part"},{"id":"T30","span":{"begin":799,"end":801},"obj":"Body_part"},{"id":"T31","span":{"begin":808,"end":810},"obj":"Body_part"},{"id":"T32","span":{"begin":883,"end":888},"obj":"Body_part"},{"id":"T33","span":{"begin":1238,"end":1251},"obj":"Body_part"},{"id":"T34","span":{"begin":1336,"end":1338},"obj":"Body_part"},{"id":"T35","span":{"begin":1655,"end":1660},"obj":"Body_part"},{"id":"T36","span":{"begin":1784,"end":1786},"obj":"Body_part"},{"id":"T37","span":{"begin":1844,"end":1853},"obj":"Body_part"},{"id":"T38","span":{"begin":1854,"end":1856},"obj":"Body_part"},{"id":"T39","span":{"begin":2061,"end":2067},"obj":"Body_part"},{"id":"T40","span":{"begin":2197,"end":2206},"obj":"Body_part"},{"id":"T41","span":{"begin":2207,"end":2209},"obj":"Body_part"},{"id":"T42","span":{"begin":2318,"end":2334},"obj":"Body_part"},{"id":"T43","span":{"begin":2335,"end":2339},"obj":"Body_part"},{"id":"T44","span":{"begin":2485,"end":2492},"obj":"Body_part"},{"id":"T45","span":{"begin":2514,"end":2516},"obj":"Body_part"},{"id":"T46","span":{"begin":2579,"end":2588},"obj":"Body_part"},{"id":"T47","span":{"begin":2589,"end":2591},"obj":"Body_part"},{"id":"T48","span":{"begin":2819,"end":2821},"obj":"Body_part"},{"id":"T49","span":{"begin":4085,"end":4090},"obj":"Body_part"},{"id":"T53","span":{"begin":4908,"end":4910},"obj":"Body_part"},{"id":"T54","span":{"begin":5016,"end":5018},"obj":"Body_part"},{"id":"T55","span":{"begin":5098,"end":5100},"obj":"Body_part"},{"id":"T56","span":{"begin":5720,"end":5722},"obj":"Body_part"},{"id":"T57","span":{"begin":5764,"end":5766},"obj":"Body_part"},{"id":"T58","span":{"begin":5797,"end":5799},"obj":"Body_part"},{"id":"T59","span":{"begin":5846,"end":5861},"obj":"Body_part"},{"id":"T60","span":{"begin":5919,"end":5921},"obj":"Body_part"},{"id":"T61","span":{"begin":6023,"end":6025},"obj":"Body_part"},{"id":"T62","span":{"begin":6238,"end":6240},"obj":"Body_part"},{"id":"T63","span":{"begin":6315,"end":6317},"obj":"Body_part"},{"id":"T64","span":{"begin":6391,"end":6396},"obj":"Body_part"},{"id":"T65","span":{"begin":6409,"end":6411},"obj":"Body_part"},{"id":"T66","span":{"begin":6453,"end":6455},"obj":"Body_part"},{"id":"T67","span":{"begin":6545,"end":6547},"obj":"Body_part"},{"id":"T68","span":{"begin":6911,"end":6913},"obj":"Body_part"},{"id":"T69","span":{"begin":7079,"end":7084},"obj":"Body_part"}],"attributes":[{"id":"A27","pred":"uberon_id","subj":"T27","obj":"http://purl.obolibrary.org/obo/UBERON_0002135"},{"id":"A28","pred":"uberon_id","subj":"T28","obj":"http://purl.obolibrary.org/obo/UBERON_0002135"},{"id":"A29","pred":"uberon_id","subj":"T29","obj":"http://purl.obolibrary.org/obo/UBERON_0002135"},{"id":"A30","pred":"uberon_id","subj":"T30","obj":"http://purl.obolibrary.org/obo/UBERON_0002135"},{"id":"A31","pred":"uberon_id","subj":"T31","obj":"http://purl.obolibrary.org/obo/UBERON_0002135"},{"id":"A32","pred":"uberon_id","subj":"T32","obj":"http://purl.obolibrary.org/obo/UBERON_0001825"},{"id":"A33","pred":"uberon_id","subj":"T33","obj":"http://purl.obolibrary.org/obo/UBERON_0000946"},{"id":"A34","pred":"uberon_id","subj":"T34","obj":"http://purl.obolibrary.org/obo/UBERON_0002135"},{"id":"A35","pred":"uberon_id","subj":"T35","obj":"http://purl.obolibrary.org/obo/UBERON_0001825"},{"id":"A36","pred":"uberon_id","subj":"T36","obj":"http://purl.obolibrary.org/obo/UBERON_0002135"},{"id":"A37","pred":"uberon_id","subj":"T37","obj":"http://purl.obolibrary.org/obo/UBERON_0001353"},{"id":"A38","pred":"uberon_id","subj":"T38","obj":"http://purl.obolibrary.org/obo/UBERON_0002135"},{"id":"A39","pred":"uberon_id","subj":"T39","obj":"http://purl.obolibrary.org/obo/UBERON_0000479"},{"id":"A40","pred":"uberon_id","subj":"T40","obj":"http://purl.obolibrary.org/obo/UBERON_0001353"},{"id":"A41","pred":"uberon_id","subj":"T41","obj":"http://purl.obolibrary.org/obo/UBERON_0002135"},{"id":"A42","pred":"uberon_id","subj":"T42","obj":"http://purl.obolibrary.org/obo/UBERON_0002494"},{"id":"A43","pred":"uberon_id","subj":"T43","obj":"http://purl.obolibrary.org/obo/UBERON_0000033"},{"id":"A44","pred":"uberon_id","subj":"T44","obj":"http://purl.obolibrary.org/obo/UBERON_0002416"},{"id":"A45","pred":"uberon_id","subj":"T45","obj":"http://purl.obolibrary.org/obo/UBERON_0002135"},{"id":"A46","pred":"uberon_id","subj":"T46","obj":"http://purl.obolibrary.org/obo/UBERON_0001353"},{"id":"A47","pred":"uberon_id","subj":"T47","obj":"http://purl.obolibrary.org/obo/UBERON_0002135"},{"id":"A48","pred":"uberon_id","subj":"T48","obj":"http://purl.obolibrary.org/obo/UBERON_0002135"},{"id":"A49","pred":"uberon_id","subj":"T49","obj":"http://purl.obolibrary.org/obo/UBERON_0000948"},{"id":"A50","pred":"uberon_id","subj":"T49","obj":"http://purl.obolibrary.org/obo/UBERON_0007100"},{"id":"A51","pred":"uberon_id","subj":"T49","obj":"http://purl.obolibrary.org/obo/UBERON_0015228"},{"id":"A52","pred":"uberon_id","subj":"T49","obj":"http://purl.obolibrary.org/obo/UBERON_0015230"},{"id":"A53","pred":"uberon_id","subj":"T53","obj":"http://purl.obolibrary.org/obo/UBERON_0002135"},{"id":"A54","pred":"uberon_id","subj":"T54","obj":"http://purl.obolibrary.org/obo/UBERON_0002135"},{"id":"A55","pred":"uberon_id","subj":"T55","obj":"http://purl.obolibrary.org/obo/UBERON_0002135"},{"id":"A56","pred":"uberon_id","subj":"T56","obj":"http://purl.obolibrary.org/obo/UBERON_0002135"},{"id":"A57","pred":"uberon_id","subj":"T57","obj":"http://purl.obolibrary.org/obo/UBERON_0002135"},{"id":"A58","pred":"uberon_id","subj":"T58","obj":"http://purl.obolibrary.org/obo/UBERON_0002135"},{"id":"A59","pred":"uberon_id","subj":"T59","obj":"http://purl.obolibrary.org/obo/UBERON_0002134"},{"id":"A60","pred":"uberon_id","subj":"T60","obj":"http://purl.obolibrary.org/obo/UBERON_0002135"},{"id":"A61","pred":"uberon_id","subj":"T61","obj":"http://purl.obolibrary.org/obo/UBERON_0002135"},{"id":"A62","pred":"uberon_id","subj":"T62","obj":"http://purl.obolibrary.org/obo/UBERON_0002135"},{"id":"A63","pred":"uberon_id","subj":"T63","obj":"http://purl.obolibrary.org/obo/UBERON_0002135"},{"id":"A64","pred":"uberon_id","subj":"T64","obj":"http://purl.obolibrary.org/obo/UBERON_0002542"},{"id":"A65","pred":"uberon_id","subj":"T65","obj":"http://purl.obolibrary.org/obo/UBERON_0002135"},{"id":"A66","pred":"uberon_id","subj":"T66","obj":"http://purl.obolibrary.org/obo/UBERON_0002135"},{"id":"A67","pred":"uberon_id","subj":"T67","obj":"http://purl.obolibrary.org/obo/UBERON_0002135"},{"id":"A68","pred":"uberon_id","subj":"T68","obj":"http://purl.obolibrary.org/obo/UBERON_0002135"},{"id":"A69","pred":"uberon_id","subj":"T69","obj":"http://purl.obolibrary.org/obo/UBERON_0000473"}],"text":"METHODS\n\nPatients\nAll adult patients (≥18 years of age) who underwent surgical intervention (n = 528) for MR due to degenerative disease between January 2004 and December 2015 at our institution were eligible for inclusion. We excluded patients with a history of previous cardiac surgery (n = 18) and patients in whom MV replacement was performed (n = 9). Of the remaining 501 patients, 17 additional patients were excluded for the following reasons: early mortality (n = 6), non-use of annuloplasty ring (n = 4) or missing predischarge echocardiogram (n = 7). The final study cohort compromised 484 patients who all underwent successful MV repair with a complete annuloplasty ring.\nMeasurements of MV gradients were acquired from continuous wave Doppler acquisitions of the diastolic inflow of the MV. Mean MV gradient was calculated as the average of 3 and 5 cycles or patients in sinus rhythm and atrial fibrillation, respectively. All acquisitions were performed according to a standardized protocol of our echo lab to ensure data reproducibility. All measurements were performed by experienced echocardiographists.\n\nStudy end points\nAll end points were defined according to the Guidelines for reporting mortality and morbidity after cardiac valve interventions [6]. Primary end points included all-cause mortality and freedom from MV reintervention. Secondary end points included freedom from atrial fibrillation. We excluded other types of atrial tachycardias (e.g. atrial flutter or incisional atrial tachycardia) because of aetiological differences associated with the development postrepair atrial tachycardias [7]. Only patients in preoperative sinus rhythm (n = 312) were included in the latter analysis.\n\nSurgical technique and perioperative care\nWhen indicated, anterior MV leaflet prolapse was addressed with chordal replacement. Posterior MV leaflet prolapse was addressed with a combination of leaflet resection and chordal replacement techniques. The decision on which technique to use was based on the extent of leaflet prolapse and excessive tissue in height and/or width. Earlier in this series, annular plication was more frequently used to help restore the continuity of the posterior MV leaflet. Later, leaflet sliding techniques were employed. Commissural prolapse was primarily addressed with papillary muscle head repositioning.\nAll patients included in the current study underwent full annuloplasty ring implantation. Ring sizing was standardly based on the surface area of the anterior MV leaflet and was not influenced by the type of anterior and/or posterior MV leaflet repair technique used. No over- or downsizing was performed in any of the patients.\nIntraoperative transoesophageal echocardiography was performed by an experienced cardiologist to document the intraoperative result of MV repair. Additionally, predischarge transthoracic echocardiography (performed on postoperative day 4–7) was performed to exclude any significant residual MR. Systolic anterior motion was observed in 2 patients and treated conservatively. Oral anticoagulation was initiated for a period of 3 months with a target Internationalized Normalized Ratio of 2.0–3.0. In presence of other indications, the target Internationalized Normalized Ratio range and treatment duration were adjusted accordingly.\n\nFollow-up\nPreoperative, intraoperative and postoperative data were retrospectively collected from our computerized database. Follow-up survival, clinical and echocardiographic data were collected through regular clinical visits at our institution or affiliated clinics and hospitals, and through questionnaires to patients. Patient follow-up was based on the available recommendations [8]. Records pertaining to reported office visits, echocardiography and rhythm information, operations, cardioversions, catheter ablation or hospitalizations were obtained and analysed. Rhythm follow-up was based on electrocardiograms obtained during regular follow-up. Additional rhythm monitoring (e.g. 24-h Holter monitoring) was performed on clinical grounds (e.g. complaints of heart palpitations) and the indication for this was left at the discretion of the attending cardiologist. A total of 1126 (mean 2.3/patient, range 0–8) follow-up echocardiograms were available for analysis.\nThe study was approved by our Institutional Ethics Committee and patient consent was obtained to allow collection and analysis of any relevant clinical and echocardiographic data. Follow-up was completed in February 2017.\n\nStatistical analysis\nContinuous data are presented as means ± standard deviation for normally distributed data or medians and interquartile range (IQR) when not normally distributed. Categorical data are presented as counts and percentages. The distribution of variables was evaluated using the Kolmogorov–Smirnov test.\nA univariable and multivariable linear regression model, with postrepair MV gradient as a dependent continuous variable, was built to explore the factors associated with postrepair MV gradient. Variable inclusion was based on the previously established effect on MV gradient. To assess for the presence of violations in model assumptions, residuals were plotted versus fitted values and investigated graphically. Univariable and multivariable Cox proportional hazards regression analysis was used to explore risk factors for the occurrence of time-to-event outcomes. Based on known clinical validity and taking into account the ratio of events to risk factors, the following factors were included in respective multivariate models: ‘for mortality’: age, gender, chronic pulmonary disease, renal impairment, left ventricular function, symptomatic MR, atrial fibrillation and postrepair MV gradient; ‘for reintervention’: anterior MV leaflet repair and postrepair MV gradient; and ‘for atrial fibrillation’: age, tricuspid valve repair, preoperative left atrial diameter and postrepair MV gradient. Cubic third-degree splines were used to adjust for possible non-linear association of mean MV gradient with time-to-event outcome in the proportional hazards model. The Wald test was used to test for the presence of non-linear effect. In addition to P-value-based hypothesis testing, non-linearity of mean MV gradient was assessed graphically by plotting the fitted splines for mean MV gradient, together with 95% confidence intervals (CIs) on the log-hazard scale versus mean MV gradient.\nTo study the evolution of mean MV gradients during the follow-up period, a mixed-model based linear regression analysis of MV gradient on follow-up time which accounts for repeated within-patient observation was carried out. To account for potential heteroscedasticity or non-normality, bootstrap-based standard errors were investigated. In addition, a post hoc exploration of model residuals was explored to investigate these same model assumptions as well as the linearity assumption of MV gradient decent versus follow-up time. No evidence of serious model deviations was found and hence the regular standard errors and hypothesis test are reported.\nAll tests were 2-tailed and a P-value of \u003c0.05 was considered statistically significant. Statistical analysis was performed using IBM Statistics for Windows, version 23.0 (IBM Corp. Released 2015. IBM SPSS Statistics for Windows, Version 23.0. Armonk, NY, USA: IBM Corp.) and R, version 3.5.1 using the Survival package, version 2.43–3 (R Foundation for Statistical Computing, Vienna, Austria)."}