Search

Background Automatic measurement of respiratory rate in general hospital patients is difficult. Patient movement degrades the signal and variation of the breathing cycle means that accurate observation for ≥60 s is needed for adequate precision. Methods We studied acutely ill patients recently admitted to a teaching hospital. Breath duration was measured from a triaxial accelerometer attached to the chest wall and compared with a signal from a nasal cannula. We randomly divided the patient records into a training (n=54) and a test set (n=7). We used machine learning to train a neural network to select reliable signals, automatically identifying signal features associated with accurate measurement of respiratory rate. We used the test records to assess the accuracy of the device, indicated by the median absolute difference between respiratory rates, provided by the accelerometer and by the nasal cannula. Results In the test set of patients, machine classification of the respiratory signal reduced the median absolute difference (interquartile range) from 1.25 (0.56–2.18) to 0.48 (0.30–0.78) breaths per min. 50% of the recording periods were rejected as unreliable and in one patient, only 10% of the signal time was classified as reliable. However, even only 10% of observation time would allow accurate measurement for 6 min in an hour of recording, giving greater reliability than nurse charting, which is based on much less observation time. Conclusion Signals from a body-mounted accelerometer yield accurate measures of respiratory rate, which could improve automatic illness scoring in adult hospital patients., A machine learning method was developed to classify sections of breathing records from acutely ill patients wearing a small wireless motion sensor. This would allow accurate and automatic measurement, recording, and charting of respiratory rate. https://bit.ly/301P8XW

Background Respiratory rate is a basic clinical measurement used for illness assessment. Errors in measuring respiratory rate are attributed to observer and equipment problems. Previous studies commonly report rate differences ranging from 2 to 6 breaths·min−1 between observers. Methods To study why repeated observations should vary so much, we conducted a virtual experiment, using continuous recordings of breathing from acutely ill patients. These records allowed each breathing cycle to be precisely timed. We made repeated random measures of respiratory rate using different sample durations of 30, 60 and 120 s. We express the variation in these repeated rate measurements for the different sample durations as the interquartile range of the values obtained for each subject. We predicted what values would be found if a single measure, taken from any patient, were repeated and inspected boundary values of 12, 20 or 25 breaths·min−1, used by the UK National Early Warning Score, for possible mis-scoring. Results When the sample duration was nominally 30 s, the mean interquartile range of repeated estimates was 3.4 breaths·min−1. For the 60 s samples, the mean interquartile range was 3 breaths·min−1, and for the 120 s samples it was 2.5 breaths·min−1. Thus, repeat clinical counts of respiratory rate often differ by >3 breaths·min−1. For 30 s samples, up to 40% of National Early Warning Scores could be misclassified. Conclusions Early warning scores will be unreliable when short sample durations are used to measure respiratory rate. Precision improves with longer sample duration, but this may be impractical unless better measurement methods are used., In acutely ill patients, the usual 30 s to count breathing are insufficient to give a reliable measurement https://bit.ly/31DNhKU

Background Anaesthetic journals frequently publish studies comparing measurement methods. A common method of analysis is the Bland and Altman plot, which relates the difference between paired measurements to the mean of the pair. Previous reviews have shown that key data are often omitted from reports using this method of analysis, and the analysis of more complex data is frequently insufficient. Methods We identified articles by searching reports, and subsequent citations, considering use of the method. We assembled a list of frequent and important criteria from these articles. These key features were tested by assessing articles in the yr 2013 and 2014, in five anaesthetic journals: Anaesthesia, Anesthesiology, Anesthesia and Analgesia, The British Journal of Anaesthesia, and The Canadian Journal of Anaesthesia. Results We found 29 features suggested for reporting such studies. Eight of these were frequently found. We chose 13 key features. In the journal articles reviewed to test these features, three features were almost always reported: the data structure, a plot of the bias, and the limits of agreement of the differences. Often, features required for adequate interpretation of the studies were absent, notably an a priori decision of acceptable limits of agreement, and an estimate of the precision of the limits of agreement. Conclusions Bland and Altman analysis remains poorly reported. Our formal list of key criteria will assist authors in providing all the relevant features of a study. We explain errors that may be made in reporting, and suggest methods for analysis, including easily available software.

A short review of the changing approach to statistics' contribution to the conduct of physiological experiments, with suggestions for further changes and better practice.

Accurate measurement of pulmonary oxygenation is important for classification of disease severity and quantification of outcomes in clinical studies. We compared predictive validity of established tension-based methods with two new measures of shunt fraction: (1) a non-invasive effective shunt (ES); and (2) inferred values from an integrated mathematical model of gas exchange (DB). Median absolute error (MAE) values for the four measures considered were: alveolar-arterial difference, 7.30kPa; PaO2/FIO2 ratio, 2.41kPa; DB, 2.13kPa; ES: 1.88kPa. ES performed significantly better than other measures (p−10 in all comparisons). While the simplicity of P/F is suitable for routine use, the superior predictive validity of ES should make this measure the preferred choice where physiological accuracy is important, such as for use as surrogate outcome in clinical research.

An amendment to this paper has been published and can be accessed via the original article.

The reliability and reproducibility of science are under scrutiny. However, a major cause of this lack of repeatability is not being considered: the wide sample-to-sample variability in the P value. We explain why P is fickle to discourage the ill-informed practice of interpreting analyses based predominantly on this statistic.

Bland and Altman have developed a measure called “limits of agreement” to assess correspondence of two methods of clinical measurement. In many circumstances, comparisons are made using several paired measurements in each individual subject. If such measurements are considered as statistically independent pairs, rather than as sets of measurements from separate individuals, limits of agreement will be too narrow. In addition, the confidence intervals for these limits will also be too narrow. Suitable software to compute valid limits of agreement and their confidence intervals is not readily available. Therefore, we set out to provide a freely available implementation accompanied by a formal description of the more advanced Bland–Altman comparison methods. We validate the implementation using simulated data, and demonstrate the effects caused by failing to take the presence of multiple paired measurements per individual properly into account. We propose a standard format of reporting that would improve analysis and interpretation of comparison studies.

Background Respiratory rate is an important measurement in patient care, but accurate measurement is often difficult. We have developed a simple non-invasive device to measure respiratory movements in clinical circumstances, with minimal interference with the patient. We investigated respiratory patterns in patients receiving postoperative morphine analgesia to assess the capacity of the device to detect abnormalities. Methods We studied subjects during self-administered opioid analgesia after major gynaecological surgery, and related the derived signals with a signal from a nasal cannula. Respiratory movement signals were transmitted wirelessly to a recorder from two encapsulated tri-axial accelerometer (RESpeck) sensors. We analysed the signals using two different sensor placements, each for 30 min. The nasal cannula signal was used to classify breathing patterns as obstructive or non-obstructed. Results We studied 20 patients for a mean duration of 49 min each. Breathing patterns were very variable, between and within patients. The median breathing rates ranged from 6.4 to 19.5 bpm. Breathing was partly obstructed in 10 patients, and six patients had repeated cycles of obstruction and transient recovery. In these patients, we found a consistent and statistically significant pattern of changes in chest wall movement, with increased abdominal and decreased rib cage movement during obstruction. In patients with slow respiratory rates, breath-to-breath times were highly variable. Conclusions In undisturbed subjects receiving patient-controlled morphine analgesia after surgery, abnormal breathing patterns are extremely common. Cyclical airway obstruction is frequent and associated with a typical pattern of changes in chest wall movement.

Key Points this final article in our series draws together some of the ideas we have addressed, and suggests important ingredients that make a paper palatable to the reviewer and the reader. #### Describe the methods. Imagine that you are starting an experiment, which involves a method that is

Key Points calvin coolidge was a taciturn president of the United States, nicknamed “Silent Cal”. When told that he had died, the acerbic writer Dorothy Parker remarked, “How could they tell?” Telling if something has not happened is a perennial problem in science, particularly so in

Opioids increase abdominal muscle activity during anaesthesia. We proposed that opioid activity during anaesthesia would change chest wall size and movement, and contribute to ventilation. Using an optical system to measure chest wall volume, we studied 10 patients during isoflurane anaesthesia, first under the influence of an opioid and then after reversal with naloxone. Measurements were made during quiet breathing and with carbon dioxide stimulation. Airway occlusion pressure was measured to assess inspiratory and expiratory muscle activity. Chest wall volume decreased with the onset of spontaneous breathing, and decreased further when breathing was stimulated by carbon dioxide. Reversal of opioid activity increased chest wall volume. Breathing movements were predominantly abdominal. Opioid action affected the timing and amplitude of breathing but the pattern of abdominal movement was not affected. Since opioids augment abdominal muscle action during expiration, the unchanged pattern of movement can be attributed to both diaphragm and abdominal activity displacing the abdominal wall reciprocally, in the inspiratory and expiratory phases of the respiratory cycle, respectively.

Scientists frequently want to answer the question ‘has this treatment had an effect?’ Most are unaware that the tests they usually use do not directly address this question. These tests usually pose a different question, based on the possibility that nothing has happened. The question becomes: ‘how probable are these data, if there were NO difference between the original populations from which the data have been randomly drawn?’ (In fact for most laboratory experiments, this supposition is patently false – the experiment has been conducted using a pre-ordained sample, possibly randomly divided into treatment and control groups, but certainly not randomly sampled.)

Key Points calvin coolidge was a taciturn president of the United States, nicknamed “Silent Cal”. When told that he had died, the acerbic writer Dorothy Parker remarked, “How could they tell?” Telling if something has not happened is a perennial problem in science, particularly so in

Key Points calvin coolidge was a taciturn president of the United States, nicknamed “Silent Cal”. When told that he had died, the acerbic writer Dorothy Parker remarked, “How could they tell?” Telling if something has not happened is a perennial problem in science, particularly so in

Background Using conscious subjects, measurement of the effects of low concentrations of anaesthetic agents can allow the dynamics of onset and offset of the agent to be measured and kinetic values estimated. However, the tests have to be rapid and preferably assess cerebral function. Methods We used a short version of the digit symbol substitution test (DSST) that allowed frequent measurement of the impairment caused by nitrous oxide. We compared 10 min of onset and offset of breathing 5% and 30% nitrous oxide in 30% oxygen, compared with 30% oxygen only. End-tidal nitrous oxide concentrations were used to predict the concentration in a central compartment, according to a range of T1/2 values chosen to be consistent with possible cerebral blood flow values. Results We studied 19 volunteers and estimated a mean response. Only 30% nitrous oxide decreased the DSST. When DSST scores were related to the values in the predicted central compartment, the best dose–effect relationship was found when the T1/2 was 37 s, consistent with a regional blood flow of about 120 ml 100 g−1 min−1. Conclusions The onset of nitrous oxide effect on DSST is rapid, consistent with the perfusion of metabolically active cerebral cortical tissues. The rate of onset is greater than previous measures based on a motor test which involved the function of subcortical structures in the central nervous system.

Scientists frequently want to answer the question ‘has this treatment had an effect?’ Most are unaware that the tests they usually use do not directly address this question. These tests usually pose a different question, based on the possibility that nothing has happened. The question becomes: ‘how probable are these data, if there were NO difference between the original populations from which the data have been randomly drawn?’ (In fact for most laboratory experiments, this supposition is patently false – the experiment has been conducted using a pre-ordained sample, possibly randomly divided into treatment and control groups, but certainly not randomly sampled.)

Key Points a jack knife is a pocket knife that is put to many tasks, because it's ready to hand. Often there could be a better tool for the job, such as a screwdriver, a scraper, or a can-opener, but these are not usually pocket items. In statistical terms, the expression implies making do with

In a previous article (Drummond & Vowler, 2012a), we discussed variation between individuals, termed variance. When measurements are taken of subjects in separate groups, the variance of the measurements within the group can be compared with the variance that exists between groups. This allows us to estimate the likelihood that groups were drawn from the same population. Commonly, a treatment has been applied to these groups, and the aim would be to find if this had increased the variance between the groups, which might then indicate that the treatment had caused an effect (Fig. 1A).

Key Points our previous article in this series considered analysis of variance ([4][1]), which is applied to measurements made of samples from different groups. That approach considers the variation of the data, and specifically the variation between samples, which is related to the presence of

Key Points • Although relatively robust to the following, ANOVA assumes • Normal distribution of residuals • Variation in each group is similar • ANOVA is sensitive to outliers, and transforms of the data may be necessary • ANOVA is a test of group differences: do at least two of the means differ from each other? • Two-way ANOVA tests a number of sources of variability • It is important to know which source of variation is of interest • Do the sources of variation interact? • Care is needed with repeated measures • There may be better tests, particularly if the factor can be expressed as a continuous variable.

Key points • Correlation and regression are used with continuous variables • Plot the variables in correlation and regression relationships to aid interpretation • An association between two discrete measurements is assessed by correlation • Regression describes and quantifies a relationship between an independent factor and a dependent variable; prediction is also possible • Few biological relationships are truly linear • Regression can be distorted by outlying values • Absence of a linear regression does not mean a relationship is not present • Regression is very frequently misused and mis-applied

Most biological scientists conduct experiments to look for effects, and test the results statistically. We have already described Student's t test which is very commonly used. However, this test concentrates on a very limited question. We assume that there is no effect in the experiment, and then estimate the possibility that we could have obtained these results. The question concerns what we may deduce, if the samples measured came from a single theoretical population of known characteristics. We calculate the probability that we might obtain the results we did, or more extreme results, on the basis of this premise. The statistical procedure is called frequentist because the results are expressed in terms of ‘how frequently would a result like this be observed, given the exact definition of the theoretical population that we proposed?’ (Fig. 1).

Most biological scientists conduct experiments to look for effects, and test the results statistically. We have already described Student's t test which is very commonly used. However, this test concentrates on a very limited question. We assume that there is no effect in the experiment, and then estimate the possibility that we could have obtained these results. The question concerns what we may deduce, if the samples measured came from a single theoretical population of known characteristics. We calculate the probability that we might obtain the results we did, or more extreme results, on the basis of this premise. The statistical procedure is called frequentist because the results are expressed in terms of ‘how frequently would a result like this be observed, given the exact definition of the theoretical population that we proposed?’ (Fig. 1).

Please cite this paper as: Drummond and Tom (2012). Presenting Data: Can You Follow A Recipe? Microcirculation 19(1), 94–98.

Please cite this paper as: Drummond and Tom (2012). Presenting Data: Can You Follow A Recipe? Microcirculation 19(1), 94–98.

Background Expiratory muscle action is prominent during anaesthesia and can impair lung function. This activity is exaggerated by the use of opioids. Airway pressure during occlusion of expiration would be a valuable measure in the study of expiratory muscle activation. However, this would only be valid if the imposed occlusion did not itself alter muscle activation. This possibility can be checked by directly assessing muscle activity by electromyography; varying arterial carbon dioxide tensions and opioid action should be considered. Methods We studied seven spontaneously breathing patients, anaesthetized with nitrous oxide and isoflurane, in four conditions: during an infusion of fentanyl and after naloxone, breathing normally and with breathing stimulated with CO2. We compared diaphragm and external oblique abdominal electromyogram (EMG) signals during normal and occluded breaths. We also measured chest wall volume and compared airway occlusion pressure, during inspiration and expiration, with the EMG results. Results Inspiratory occlusion increased the duration of inspiration during hypercapnia by 20%, but not the rate of electrical activation of the diaphragm, indicating that occlusion does not cause a reflex increase in diaphragm contraction. In contrast, expiratory occlusion did not affect either the duration of expiration or the electrical activity of the external oblique muscles. Conclusions In these conditions, except for a change in inspiratory duration, respiratory muscle activity is unaffected by airway occlusion. Airway occlusion will permit valid measures of muscle activity in inspiration and expiration and provide simple measurements of respiratory muscle function during anaesthesia.

Please cite this paper as: Drummond and Tom (2012). Presenting Data: Can You Follow A Recipe? Microcirculation 19(1), 94–98.

• ‘Student’ was a statistician who worked in quality control. • The t-test asks ‘how probable are these samples, if they have been sampled from the same source?’ • Considering what would be found when repeated samples are taken is the frequentist approach to statistical testing. • Student’s t-test does not indicate the probability that the null hypothesis is true. • Other methods of testing experimental results can be more appropriate. • We may need to ask ‘how different?’ or ‘is there no effect?’

Background Respiratory rate is an important measurement in patient care but frequently poorly assessed. We set out to develop a simple non-invasive device to reliably measure respiratory movements and estimate respiratory rate, in clinical circumstances. Methods Respiratory movement was detected with an encapsulated tri-axial accelerometer (Orient speck) and the data transmitted wirelessly to a computer for analysis. We studied subjects after gynaecological surgery who received opioid analgesia, and compared the derived signal with a signal from nasal cannula using directly matched breaths and within the same 5 min epoch. We analysed the signals for 5 min epochs over a 15 h recording period. Results For matched breath analysis, the instantaneous respiratory rates matched within 2 bpm on 86% of occasions. A similar match was found between epoch averages of the respiratory rate. The mean absolute difference between the respiratory rate measured by nasal cannula and Orient speck was 0.6 bpm. The Orient speck generated reliable measures of respiratory rate every 5 min in 95.4% of epochs. Conclusions The Orient speck provides a reliable measure of respiratory rate at frequent intervals in subjects receiving patient-controlled morphine analgesia after surgery.

• Samples are often compared by first proposing that they could have come from the same population • Two random samples from the same population are unlikely to be the same • Although not exactly the same, is it probable that two samples have been drawn from the same population? • If the probability they came from the same population is small, then the original proposal is not supported • If it is improbable that the difference between the mean values of the two samples could be zero, this is poor support for the proposal that the samples have come from the same source • A ‘confidence interval’ is the range of values that could contain the true value • Small samples are often imprecise, so a difference between samples may be concealed • Precision is also affected by the variability in a population: this could swamp the ‘signal’ difference • Significant does not mean important, only unlikely to have happened by chance

Email: g.b.drummond@ed.ac.ukExperimental dataareanalysedstatisticallyto allow us to draw conclusions from alimited set of measurements. The hardfact is that we can never be certain thatmeasurements from a sample will exactlyreflect the properties of the entire group ofpossible candidates available to be studied(although using a sample is often the onlypractical thing to do). It’s possible thatsome scientists are not even clear that theword ‘sample’ has a special meaning instatistics, or understand the importanceof taking an

Summary Invasive ventilation poses a risk of respiratory infection that can be drug-resistant. One means of reducing transmission of infection is the use of a breathing system filter. Filters are intended to be used with dry gas. Current international standards do not require that filters prevent bacterial transfer when wet. It is not known whether microorganisms pass through wet filters, but theory predicts that this might occur. We tested six filters from three different manufacturers. We passed a suspension of microorganisms through the filters using the least pressure necessary, and incubated a sample of the filtrate on blood agar. All the filters tested allowed free passage of both Candida albicans and coagulase-negative staphylococci. The median (IQR [range]) pressure required for fluid to flow across the filter varied greatly between different filter types (20 (0–48 [0–138]) cmH2O). We conclude that even large microorganisms pass across moist breathing system filters in conditions that are found in clinical practice.

Objective To compare recovery from sedation using remifentanil and propofol with our standard regimen of morphine and midazolam for closed reduction in shoulder dislocation in an ED. Methods We randomised 40 patients for closed shoulder reduction to receive either remifentanil and propofol (R&P) (20 patients) or morphine and midazolam (M&M) (20 patients). A suitably trained ED doctor gave the sedation. R&P patients received oxygen from a tight-fitting facemask and Mapleson C circuit to prevent hypocapnia and thus reduce the risk of apnoea. Shoulder reduction was performed by another doctor with an ED nurse also present. We noted recovery times, pain and sedation scores, operative conditions, additional medication and adverse events. Results All patients given remifentanil and propofol had recovered within 30 min in contrast to the morphine and midazolam group where 17 of 20 patients had recovered after 60 min, the remainder requiring a total of 90 min. Median recovery times were 15 min (95% CI 15 to 20) for the remifentanil and propofol group, and 45 min (95% CI 29 to 48) for the morphine and midazolam group. Reduction conditions and scores for pain/distress did not differ between the groups. Seven patients required additional sedation (four R&P, three M&M) to enable shoulder reduction. Five patients (three R&P, two M&M) had received analgesia prior to the procedure from the ambulance service (one R&P, one M&M) and ED (two R&P). Two patients given morphine and midazolam required flumazenil to counter oversedation. Conclusions Remifentanil and propofol reduces patient recovery time and provides equivalent operating conditions compared with morphine and midazolam for the reduction of anterior glenohumeral dislocation.

Editor,We present two examples in which a combination of two items of equipment, made by different manufacturers, caused unforeseen errors. Both involved capnography, which is a routine monitoring requirement in many countries.Standard connections for anaesthetic equipment allow products from differ

Reporting of ethical matters in The Journal is very important. To advise and assist authors, particularly those who may be less familiar with the legislation in the UK, this article sets out the basic principles and methods that should be used and provides many key web sources of information. It addresses the structure of regulations, and introduces the concept of research governance. The UK law is summarized. Advice is given on the format and description of experiments, and common problems addressed. Aspects of human studies are addressed. Ethical considerations of publication such as authorship and originality, and problems such as plagiarism and fabrication are described. Updates will be published regularly.

Background and objective In patients after surgery, we observed large-amplitude low-frequency changes in digital plethysmograph measurements when DC coupling of the signal was used. We set out to assess factors that might contribute to these events and in particular to test the possibility that low-frequency signals could be used to assess respiratory disturbances. Methods We recorded values in 23 patients who had undergone gynaecological surgery. We measured nasal flow, abdominal pressure (by urinary catheter), venous pressure in the hand, and DC-coupled optical transmission plethysmography. Signals were replayed and analysed to assess the incidence of specific patterns of events. Results Most patients received morphine for postoperative analgesia. Respiratory irregularity and expiratory muscle action were very frequent. Increases in abdominal pressure during expiration caused increases in venous pressure and pulsation. In 12 out of 23 patients, a characteristic response consistent with vasoconstriction was noted after increases in breath size, and, in seven patients, very-low-frequency (0.2–0.7 Hz) oscillations of finger volume were present that appeared unrelated to respiratory events. Patients who did not receive morphine had very different plethysmograph patterns, with significantly smaller pulse amplitude. Conclusion Low-frequency changes in finger volume can be simply obtained and provide considerable information about peripheral circulatory dynamics. Diverse patterns can be recognized, but the range of responses suggests that current techniques cannot be used alone to assess cardiorespiratory status. However, a combination of plethysmography with respiratory measurements shows characteristic events.

The constant SCO2 (solubility factor 1⁄4 0.03 mm litre mm Hg or 0.23 mm litre kPa) allows conversion of PCO2 into [CO2]. Acid–base disturbances could now be classified as respiratory or non-respiratory but this had limitations. For pH to be determined by CO2 and bicarbonate, these two had to be independent of each other, which is not possible. In addition, the equation did not take into account non-volatile weak acids present in the body. To overcome these limitations, Siggard-Anderson proposed PCO2-independent indicators of acid–base disturbances. These include standard bicarbonate, buffer base and base excess/deficit. However, these approaches do not take into account PCO2 and do not differentiate between different causes of nonrespiratory acid–base disturbances. In the 1980s, Stewart proposed a physicochemical approach to acid–base physiology. The complex calculations involved in this approach were not possible in the precomputer era but are easily done today. He wrote of this approach having ‘revolutionized our ability to understand, predict and control what happens to hydrogen ions in living systems’. Definitions