FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2620123073> ?p ?o ?g. }

• W2620123073 endingPage "949" @default.
• W2620123073 startingPage "948" @default.
• W2620123073 abstract "To the Editor:We read with interest the study by Bernstein et al.,1Bernstein G Mannino FL Heldt GP Callahan JD Bull DH Solo A et al.Randomized multicenter trial comparing synchronized and conventional intermittent mandatory ventilation in neonates.J Pediatr. 1996; 128: 453-463Abstract Full Text Full Text PDF PubMed Scopus (106) Google Scholar who described the benefits of synchronized intermittent mandatory ventilation (SIMV) over conventional intermittent mandatory ventilation (IMV) in neonates. SIMV has been tried in the past for weaning infants from the ventilators, with equivocal results.2Dimitriou G Greenough A Griffin F Chan V Synchronous intermittent mandatory ventilation modes compared with patient-triggered ventilation during weaning.Arch Dis Child. 1995; 72: F188-F190Crossref Google Scholar We have some questions regarding the study design, data analysis, and conclusions. Inclusion criteria include abnormal lung parenchyma on chest x-ray film, which is not clearly defined and could be extremely subjective. Is the period of ventilation on high-frequency ventilation included in the duration of ventilation? The study has included periods of reintubation in the duration of ventilation, which can change the results significantly because the reason for reintubation could be entirely unrelated to the initial problem.There is no basis for combining 0.1 mg of morphine with 0.1 mg of any benzodiazepine and 1 μg of fentanyl: the use of analgesics or sedatives in the neonatal intensive care unit is subjective. Pain or comfort scores, if done, could be more appropriate. There is no mention of criteria for use of surfactant (prophylactic vs rescue), antenatal steroids, indications for extubation and reintubation, and extubation failures in both groups.The statistical models and methods used in the analysis are questionable. Measurements were taken at 16 time points, ranging from 30 minutes to 240 hours, but the results were reported only at 30 minutes and 1 hour if significant for some of the response variables. Are the specific tests in the analysis performed by design, or are they data directed? If the latter is the case, which it appears to be, then it invalidates all the results of the analysis. What are the clinical implications of finding something significant at 30 minutes or 1 hour, but nothing significant at any other time point?Another critical issue is the multiple testing. When multiple tests are performed, it is important to adjust the significance level of each test, or at least use a more stringent level of significance, such as 0.01. The p values reported as p less than 0.05 are very close to 0.05, and considering the total number of tests performed, it is difficult to justify statistical significance for these marginal cases. Even when a statistical significance is observed, one should ask whether it is clinically significant. The authors found statistical significance in the difference in mean airway pressure for all infants at 1 hour on the basis of the sample means of 8.08 and 8.63 (p = 0.04). Is this difference clinically significant? In fact, if one uses the rounded numbers 8.1 and 8.6, and with this sample size, the difference is not even statistically significant (p = 0.06).As mentioned earlier, the authors took measurements at 16 different points and performed statistical tests at each (or some) of these points. For any meaningful inference, some form of longitudinal analysis using repeated measures should have been employed. Rather than classifying infants into different birth weight (BW) groups and performing a separate analysis for each group, it would have been better to use BW as a covariate and regress out its effect, and to compare the two treatments after adjustment for the BW effect. Similarly, the center effect and other relevant factors, such as age, sex, and Apgar scores, could have been used as covariates. How can the log-rank test be used to compare duration of mechanical ventilation or the age to regain BW for the survivors3Dawson-Saunders B Trapp RG Basic and clinical biostatistics.in: 2nd ed. : Appleton & Lange, New York1994: 200-209Google Scholar? The log-rank test is a test for comparing two (or more) survival functions. The authors mentioned using 5% confidence interval with 90% power, which we hope is a typographical error. It is mentioned in the abstract that 3 of 46 infants receiving IMV and none of the 47 infants receiving SIMV required extracorporeal membrane oxygenation (ECMO) (no p value given). Can SIMV prevent infants from qualifying for ECMO? This study, in summary, is nonmasked, and negative results disapproving the hypothesis have not been mentioned.9/35/77123 To the Editor:We read with interest the study by Bernstein et al.,1Bernstein G Mannino FL Heldt GP Callahan JD Bull DH Solo A et al.Randomized multicenter trial comparing synchronized and conventional intermittent mandatory ventilation in neonates.J Pediatr. 1996; 128: 453-463Abstract Full Text Full Text PDF PubMed Scopus (106) Google Scholar who described the benefits of synchronized intermittent mandatory ventilation (SIMV) over conventional intermittent mandatory ventilation (IMV) in neonates. SIMV has been tried in the past for weaning infants from the ventilators, with equivocal results.2Dimitriou G Greenough A Griffin F Chan V Synchronous intermittent mandatory ventilation modes compared with patient-triggered ventilation during weaning.Arch Dis Child. 1995; 72: F188-F190Crossref Google Scholar We have some questions regarding the study design, data analysis, and conclusions. Inclusion criteria include abnormal lung parenchyma on chest x-ray film, which is not clearly defined and could be extremely subjective. Is the period of ventilation on high-frequency ventilation included in the duration of ventilation? The study has included periods of reintubation in the duration of ventilation, which can change the results significantly because the reason for reintubation could be entirely unrelated to the initial problem.There is no basis for combining 0.1 mg of morphine with 0.1 mg of any benzodiazepine and 1 μg of fentanyl: the use of analgesics or sedatives in the neonatal intensive care unit is subjective. Pain or comfort scores, if done, could be more appropriate. There is no mention of criteria for use of surfactant (prophylactic vs rescue), antenatal steroids, indications for extubation and reintubation, and extubation failures in both groups.The statistical models and methods used in the analysis are questionable. Measurements were taken at 16 time points, ranging from 30 minutes to 240 hours, but the results were reported only at 30 minutes and 1 hour if significant for some of the response variables. Are the specific tests in the analysis performed by design, or are they data directed? If the latter is the case, which it appears to be, then it invalidates all the results of the analysis. What are the clinical implications of finding something significant at 30 minutes or 1 hour, but nothing significant at any other time point?Another critical issue is the multiple testing. When multiple tests are performed, it is important to adjust the significance level of each test, or at least use a more stringent level of significance, such as 0.01. The p values reported as p less than 0.05 are very close to 0.05, and considering the total number of tests performed, it is difficult to justify statistical significance for these marginal cases. Even when a statistical significance is observed, one should ask whether it is clinically significant. The authors found statistical significance in the difference in mean airway pressure for all infants at 1 hour on the basis of the sample means of 8.08 and 8.63 (p = 0.04). Is this difference clinically significant? In fact, if one uses the rounded numbers 8.1 and 8.6, and with this sample size, the difference is not even statistically significant (p = 0.06).As mentioned earlier, the authors took measurements at 16 different points and performed statistical tests at each (or some) of these points. For any meaningful inference, some form of longitudinal analysis using repeated measures should have been employed. Rather than classifying infants into different birth weight (BW) groups and performing a separate analysis for each group, it would have been better to use BW as a covariate and regress out its effect, and to compare the two treatments after adjustment for the BW effect. Similarly, the center effect and other relevant factors, such as age, sex, and Apgar scores, could have been used as covariates. How can the log-rank test be used to compare duration of mechanical ventilation or the age to regain BW for the survivors3Dawson-Saunders B Trapp RG Basic and clinical biostatistics.in: 2nd ed. : Appleton & Lange, New York1994: 200-209Google Scholar? The log-rank test is a test for comparing two (or more) survival functions. The authors mentioned using 5% confidence interval with 90% power, which we hope is a typographical error. It is mentioned in the abstract that 3 of 46 infants receiving IMV and none of the 47 infants receiving SIMV required extracorporeal membrane oxygenation (ECMO) (no p value given). Can SIMV prevent infants from qualifying for ECMO? This study, in summary, is nonmasked, and negative results disapproving the hypothesis have not been mentioned. We read with interest the study by Bernstein et al.,1Bernstein G Mannino FL Heldt GP Callahan JD Bull DH Solo A et al.Randomized multicenter trial comparing synchronized and conventional intermittent mandatory ventilation in neonates.J Pediatr. 1996; 128: 453-463Abstract Full Text Full Text PDF PubMed Scopus (106) Google Scholar who described the benefits of synchronized intermittent mandatory ventilation (SIMV) over conventional intermittent mandatory ventilation (IMV) in neonates. SIMV has been tried in the past for weaning infants from the ventilators, with equivocal results.2Dimitriou G Greenough A Griffin F Chan V Synchronous intermittent mandatory ventilation modes compared with patient-triggered ventilation during weaning.Arch Dis Child. 1995; 72: F188-F190Crossref Google Scholar We have some questions regarding the study design, data analysis, and conclusions. Inclusion criteria include abnormal lung parenchyma on chest x-ray film, which is not clearly defined and could be extremely subjective. Is the period of ventilation on high-frequency ventilation included in the duration of ventilation? The study has included periods of reintubation in the duration of ventilation, which can change the results significantly because the reason for reintubation could be entirely unrelated to the initial problem. There is no basis for combining 0.1 mg of morphine with 0.1 mg of any benzodiazepine and 1 μg of fentanyl: the use of analgesics or sedatives in the neonatal intensive care unit is subjective. Pain or comfort scores, if done, could be more appropriate. There is no mention of criteria for use of surfactant (prophylactic vs rescue), antenatal steroids, indications for extubation and reintubation, and extubation failures in both groups. The statistical models and methods used in the analysis are questionable. Measurements were taken at 16 time points, ranging from 30 minutes to 240 hours, but the results were reported only at 30 minutes and 1 hour if significant for some of the response variables. Are the specific tests in the analysis performed by design, or are they data directed? If the latter is the case, which it appears to be, then it invalidates all the results of the analysis. What are the clinical implications of finding something significant at 30 minutes or 1 hour, but nothing significant at any other time point? Another critical issue is the multiple testing. When multiple tests are performed, it is important to adjust the significance level of each test, or at least use a more stringent level of significance, such as 0.01. The p values reported as p less than 0.05 are very close to 0.05, and considering the total number of tests performed, it is difficult to justify statistical significance for these marginal cases. Even when a statistical significance is observed, one should ask whether it is clinically significant. The authors found statistical significance in the difference in mean airway pressure for all infants at 1 hour on the basis of the sample means of 8.08 and 8.63 (p = 0.04). Is this difference clinically significant? In fact, if one uses the rounded numbers 8.1 and 8.6, and with this sample size, the difference is not even statistically significant (p = 0.06). As mentioned earlier, the authors took measurements at 16 different points and performed statistical tests at each (or some) of these points. For any meaningful inference, some form of longitudinal analysis using repeated measures should have been employed. Rather than classifying infants into different birth weight (BW) groups and performing a separate analysis for each group, it would have been better to use BW as a covariate and regress out its effect, and to compare the two treatments after adjustment for the BW effect. Similarly, the center effect and other relevant factors, such as age, sex, and Apgar scores, could have been used as covariates. How can the log-rank test be used to compare duration of mechanical ventilation or the age to regain BW for the survivors3Dawson-Saunders B Trapp RG Basic and clinical biostatistics.in: 2nd ed. : Appleton & Lange, New York1994: 200-209Google Scholar? The log-rank test is a test for comparing two (or more) survival functions. The authors mentioned using 5% confidence interval with 90% power, which we hope is a typographical error. It is mentioned in the abstract that 3 of 46 infants receiving IMV and none of the 47 infants receiving SIMV required extracorporeal membrane oxygenation (ECMO) (no p value given). Can SIMV prevent infants from qualifying for ECMO? This study, in summary, is nonmasked, and negative results disapproving the hypothesis have not been mentioned. 9/35/77123" @default.
• W2620123073 created "2017-06-05" @default.
• W2620123073 creator A5020818251 @default.
• W2620123073 creator A5042429013 @default.
• W2620123073 creator A5090105154 @default.
• W2620123073 date "1996-12-01" @default.
• W2620123073 modified "2023-09-23" @default.
• W2620123073 title "Randomized multicenter trial comparing synchronized and conventional intermittent mandatory ventilation in neonates" @default.
• W2620123073 cites W1552511798 @default.
• W2620123073 cites W2004236406 @default.
• W2620123073 cites W2055242080 @default.
• W2620123073 doi "https://doi.org/10.1016/s0022-3476(96)70058-6" @default.
• W2620123073 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/8969751" @default.
• W2620123073 hasPublicationYear "1996" @default.
• W2620123073 type Work @default.
• W2620123073 sameAs 2620123073 @default.
• W2620123073 citedByCount "0" @default.
• W2620123073 crossrefType "journal-article" @default.
• W2620123073 hasAuthorship W2620123073A5020818251 @default.
• W2620123073 hasAuthorship W2620123073A5042429013 @default.
• W2620123073 hasAuthorship W2620123073A5090105154 @default.
• W2620123073 hasBestOaLocation W26201230731 @default.
• W2620123073 hasConcept C126322002 @default.
• W2620123073 hasConcept C127413603 @default.
• W2620123073 hasConcept C168563851 @default.
• W2620123073 hasConcept C177713679 @default.
• W2620123073 hasConcept C187212893 @default.
• W2620123073 hasConcept C194828623 @default.
• W2620123073 hasConcept C200457457 @default.
• W2620123073 hasConcept C71924100 @default.
• W2620123073 hasConcept C78519656 @default.
• W2620123073 hasConceptScore W2620123073C126322002 @default.
• W2620123073 hasConceptScore W2620123073C127413603 @default.
• W2620123073 hasConceptScore W2620123073C168563851 @default.
• W2620123073 hasConceptScore W2620123073C177713679 @default.
• W2620123073 hasConceptScore W2620123073C187212893 @default.
• W2620123073 hasConceptScore W2620123073C194828623 @default.
• W2620123073 hasConceptScore W2620123073C200457457 @default.
• W2620123073 hasConceptScore W2620123073C71924100 @default.
• W2620123073 hasConceptScore W2620123073C78519656 @default.
• W2620123073 hasIssue "6" @default.
• W2620123073 hasLocation W26201230731 @default.
• W2620123073 hasLocation W26201230732 @default.
• W2620123073 hasOpenAccess W2620123073 @default.
• W2620123073 hasPrimaryLocation W26201230731 @default.
• W2620123073 hasRelatedWork W2052539084 @default.
• W2620123073 hasRelatedWork W2235574018 @default.
• W2620123073 hasRelatedWork W2507295142 @default.
• W2620123073 hasRelatedWork W2953633878 @default.
• W2620123073 hasRelatedWork W3151010708 @default.
• W2620123073 hasRelatedWork W3159250744 @default.
• W2620123073 hasRelatedWork W3195440323 @default.
• W2620123073 hasRelatedWork W4256514411 @default.
• W2620123073 hasRelatedWork W4292236216 @default.
• W2620123073 hasRelatedWork W2083697902 @default.
• W2620123073 hasVolume "129" @default.
• W2620123073 isParatext "false" @default.
• W2620123073 isRetracted "false" @default.
• W2620123073 magId "2620123073" @default.
• W2620123073 workType "article" @default.