shallow breathing index as an indicator of patient work of breathing during pressure support ventilation

shallow breathing index as an indicator of patient work of breathing during pressure support ventilation

Use of the rapid/shallow breathing as an indicator of patient work of breathing during pressure support ventilation Jay A. Johannigman, Fred A. Luchet...

1MB Sizes 0 Downloads 62 Views

Use of the rapid/shallow breathing as an indicator of patient work of breathing during pressure support ventilation Jay A. Johannigman, Fred A. Luchette,

MD, MD,

Kenneth a& James

Davis Jr, MD, M. Hurst, MD,

Robert

S. Campbell, Ohro

RRT,

Richard

D. Branson,

index

RRT,

C277c27277nt2,

Bachgw.ntnd. il~eeasul~ng-patierlt ujork of breathing (WOB J has been suggested to provide safe, agpessive weuning from mechanical ventilation. I’ve compared PF I OBpt and przssu7-e~timejrroduct (PTP) to routine weaning parametexs [breath rate If), tidal volume (I,‘$, frequency/tidml volume ratio (f/I,;)] at diJferen t le-uelsof p lessure ^ suppol t ventilation (PSI:). Methods. Fifteen patients in the surgical intensizle care unit requiCng prolonged weaning (more than 3 da!s) ulere entered in th,estudy A balloon-tipped esophageal catheter was placed and position confirmed b! tnspection of pressfi,re and flow xfazreforms. Each patie& was Iandomb assigned to breathe with 5, 10, I?, and 20 cm HZ0 of PSI,: After 30 minutes, 40 breaths were recorded and anatjxed. ~l~feasurmen t of WOBpt, PTe J; I,:, and f/I:rT zL!en?made uszng the Btcore CP-100 monitor: Mean values for each parameter were cakxlated. PTP and WOBp, were plotted against f/I', to determine correlation coefficient. Results. PTe 1170Bl,t,and f/I) decreased in a stepwise fashion us PSI’ was inci-eased. The f/VT correlated most closel\l mth WOBp, (r = 0.983) and PTP (r = 0.972). Monitoring f alone also correlated with WOB,, (r = 0.894) and PTP (r = 0.881). .-Ill patients ulere weaned from the ventilator (mean duration, 5 2 +-- J. 9 day). AVinepatients required tracheostomq beforefinal liberation from the ventilator Conclusions. Direct measurement of WOBF, 5 invasive, expensizze, and maJ be confusing to clirzicians. hIonitori?2gf/T; may be u.seful when changing PST~~during weaning. (Surgery 1997;122:737-41. j

iik.ANING

FROM

XIECHANICV.

VENTILATION

invOh?S

the gradual withdrawal of mechanical ventilator-) support and the transfer of mechanical work of breathing from the ventilator to the patient’s respiratory muscles. Successful weaning requires the timely withdrawal of mechanical ventilatory assistance in an attempt to avoid unnecessary, prolonged ventilation. This process must also balance and maintain patient comfort and safety while avoiding respiraton; muscle fatigue. Debate continues among clinidians as to the best method of weaning mechanical ventilation’.’ and the most useful weaning parameters that mav be used to Presented at the Ftftv-fourth Surgical Association, Chicago.

.dnnual

of the Central 7-9. 1997.

Meeting

Ill., March

Reprint requests’ Jay A. Johannigman, MD. Trauma~‘CritIcal Care, Department of Surgery, Cincinnati College of Medicine, 231 Bethesda 2457), Cincinnati, OH 45267-0555. Copyright 0039-606OiQ7

@ 1997 Mosbv-Year ‘$5.01) + 0

Book, 11/6/83430

Inc.

Divtsion of University of -Are. (Room

predict success and measure patient response.3-6 The rapid/shallow breathing index, also known as the frequency to tidal volume ratio (f/VT), was first described as a weaning and extubation parameter by Yang and Tobin in 1990. They reported that a f/V, of greater than 105 during unassisted breathing was the most accurate predictor of weaning failure in their patients. The reliability of this index has been called into question by others who have attempted to duplicate their work, although in different patients5 and with various levels of ventilatory support. s Still others have suggested that invasivelv monitoring WOBpt may be useful in decreasing the \t:eaning time while providing increased patient comfort and more appropriate respiratory muscle unloading during the weaning process. ’ l1 The purpose of this study was to evaluate the effectiveness of using the f/VT as an estimator of WOB,, during spontaneous breathing with various levels of pressure support ventilation (PSV) SURGERY

737

738

Pwa

Johannigman

PS”&

et al.

PS”.IO

surgery October 1997

PS”.IS

PJ”-m

Fig. 1. The relationship between f/VT (bpm/L) and X 100) as pressure support level is WOBp, (joules/L increased.

Fig. 2. The relationship between f/VT (bpm/L) and PTP (cm H,O/s/m) as pressure support level is increased.

METHODS Fifteen mechanically ventilated patients recovering from acute respiratory failure (ARF) in the surgical intensive care unit were studied. The etiology was trauma (8), pneumonia (4), or postoperative (3). Each patient was considered to have partial resolution of the underlying indication for mechanical ventilation and active weaning was in progress. Patients were enrolled into the study if liberation from mechanical ventilation was not accomplished within 14 days or active weaning attempts had proved unsuccessful for more than 72 hours. Patient work of breathing (joules/L), pressuretime-product (cm H*O/s/m), frequency (f) , tidal volume (V,), and the index of rapid/shallow breathing (f/V,) were measured using the CP-100 pulmonary monitor (Allied Healthcare Products, Riverside, CA). An esophageal balloon catheter was placed into the distal third of the esophagus, and a variable orifice flow transducer wasplaced between the endotracheal tube and breathing circuit. Proper placement of the esophageal catheter was

confirmed using the technique described by Baydur et al.l* All patients were breathing spontaneously with 5 cm H,O continuous positive airway pressure (CPAP) and between 5 and 20 cm H,O PSV. Flow triggering was used, if available; otherwise the pressure sensitivity was set to the most sensitive setting possible to avoid autotriggering. A standard 72inch disposable breathing circuit was used (Hudson RCI, Temecula, CA) with a heated humidifier (MR730, Fischer-Paykel, Auckland, New Zealand). Each patient was intubated with either a size 7.5 or 8.0 mm endotracheal tube (Mallinckrodt Medical, St. Louis, MO ) at the time of measurement. Each patient was assigned to each of the following CPAP/PSV combinations (5-CPAP/O-PSV, 5CPAP/5-PSV, 5-CPAP/lO-PSV, 5-CPAP/l5-PSV, 5CPAP/20-PSV) in random order. After a 30 minute period of stabilization at each setting, 40 consecutive breaths were recorded and averaged. “Acceptable work” was arbitrarily defined, based on previously published work, for each of the measured parameters as follows: f < 30, WOB,, < 1.0 joules/L, f/VT < 105, and PTP < 100. The f/VT was compared to the WOB,, (joules/L) and PTP at each level of PSV. RESULTS Table I shows the mean values for each of the measured parameters at all levels of PSV. The f/VT correlated most closely with invasively measured WOB t (r= 0.983). The f/VT also closely correlated with i: TP (r = 0.972). Frequency alone correlated well with WOB,, (r = 0.894) and PTP (r = 0.881). The relationship between f/VT and WOB,, is illustrated in Fig. 1. The relationship between f/VT and PTP is illustrated

in Fig. 2. All measured

parame-

ters (f, PTP, f/V,, WOB,,) decrease in a stepwise fashion as PSV is increased. Exhaled V, increased incrementally with each increase in PSV. A PSV level of 5 cm H,O was necessary to provide “acceptable work” using the f/VT end point. A PSV level of 10 cm H,O provided an acceptable work of breathing using all other measured parameters asend points. Two patients had f > 30 at the highest setting of 20 PSV. One other patient had WOB t > 1.0 and a PTP > 100 at 20 PSV. Table II revea fs the number of patients with “acceptable” work based on each of the measured variables. All patients were eventually liberated from the ventilator (duration of mechanical ventilation = 22 + 6 days). Nine patients required tracheotomy before final discontinuation from the mechanical ventilator.

Johann&man

et al. 739

Table I. Values (mean + SD) for f. V,, WOB t, and PTP at each of the setting combinations Setting

5CPW

f @pm!

30.1 * 11.4 221.2f 62.6 107.1f 86.5 1.14+ 0.466

PTP (cm H,O/‘s/min) f/VT (bpmPL) WOB I (joules/L)

O-PSV

5.CPL;IP 5.PSV

31.1 * 7.9 167.8f 73.3 90.8 f 44.7 0.89 + 0.52

5-w;Ip 27.7 94.2 68.5 0.58

1 o-PSV

5-CMP

15-PSV

5-c&w

20.PST’

f 7.2 xi 74.8 f 38.6

21.8 + 7.6

18.5 f 8

58.6 + 60.6 44.3 + 24.6

+ 0.61

0.41 * 0.5

47.2 f 35.6 33.8 + 23.3 0.32 AZ0.40

Table II. Number of patients displaying “acceptable work” as determined by each of the measured parameters at every level of ventilatory support (PSV) tested Setting

f f/&T M:OBpt WOBpt.min PTP

5-CPLkP O-PSV

10 (66%) 10 (66%) 4 (26%) 4 (26%)

0 (0%)

.5-GRIP 5-PSV

S-GEM 1 0-PSV

9 (60%)

10 (66%) 12 (80%) 10 (66%) 12 (80%)

11 (73%) 8 (53%) 7 (46%) 3 (20%)

DISCUSSION The level of ventilatory support during weaning should be titrated continually to hasten the process, maintain patient safety, and avoid fatigue of the respiratory muscles. The concept of measuring WOBp, during weaning is not new, but remains puzzling to clinicians. The technique for measuring WOB,, was first described in 1950 by Otis et a1.13The work of breathing required in normal volunteers has been previously reported.14,15 The threshold value for work of breathing (above which mechanical ventilation is necessary) has been reported to be between 13.4 and 18.0 joules /L/min.16a17 Thus the “acceptable” work of breathing for each patient during weaning is somewhere between 0.6 and 1.34 joules/L. This level will vary between patients and will likely vary for the same patient during his/her course on the mechanical ventilator, as inherent resistance and compliance characteristics change. The work of breathing may be classified as mechanical or metabolic. The definition of work is the product of the force in the direction of movement times the distance the resisting object is moved.18 Applying this definition to work of breathing is difficult, as the respiratory muscles may expend substantial amounts of energy with minimal or no movement of gas volume. Measurement of the metabolic work of breathing may be more meaningfL~1clinically than mechanical measures.T~vo methods of measuring the metabolic work of the respiratory muscles include the oxygen cost of breathing6 and pressure-time-product.lg The energy cost of the respiratory muscles is proportional to th e tension, duration, velocity, and

7 (46%))

5-GNP

15-PSV

5-GNP

20.PSV

13 (86%)

14 (93%)

15 (100%)

15 (100%)

12 (80%) 13 (86%) 12 (80%)

14 (93%) 14 (93%) 14 (93%)

frequency of contractions. The PTP accounts for all of these components except velocity. For this reason, we believe the PTP to be a superior indicator of patient work of breathing clinically. The change in f/VT in our patients in response to varying the PSV level correlated closely with both the WOB,, and PTP. Measuring the f/VT is easily and reliably accomplished and should be familiar to clinicians caring for mechanically ventilated patients. In our opinion, this makes monitoring of f/VT attractive to aid clinicians in determining the optimal level of and patient response to PSV. Banner et al.‘O studied the relationship between breathing frequency, breathing pattern, and patient work of breathing and concluded that WOBp, should be measured directly because the f/VT may be “inaccurate” or “misleading” in inferring WOBp,. Although this may be true in terms of reflecting a certain number, our results suggest that the actual change in f/V,, WOB,,,, and PTP are closely correlated in response to varying the pressure support level. We are currently unaware of any comparative clinical study that has shown that routinely monitoring WO$, and maintaining it within a certain range will result in more efficient, successful,or safer weaning from mechanical ventilation. A recent study by Gluck et al.‘O set out to determine whether using invasive measurements obtained from esophageal manometry would result in more aggressiveweaning. They concluded that use of this invasive information decreasedweaning time an average of 1.7 daysfaster than use of standard weaning indices. This study was poorly designed, however, and makes assumptions based on their weaning protocol and institutional

740 Johannigman

et al.

technique of making only one weaning attempt (incremental change in ventilation) per day. We believe that titration of mechanical ventilator) support is a better term and technique than weaning ventilatory support. Titration implies that clinicians are continually assessing the patient’s need for and response to mechanical ventilation. This allows adjustments to be made regularly in response to changes in patient condition as opposed to weaning patients until they have failed the weaning trial based on one or more clinical indicators before increasing the level of ventilatory support. Our results demonstrate that monitoring F/VT provides clinically relevant information regarding patient response to manipulating the PSV level. Further work is necessary to determine the impact of monitoring f/V, on weaning time, prediction of successful extubation. and outcome.

CONCLUSION Direct measurement of WOQ is invasive and expensive and, if not familiar to the clinician, ma)r be conf&ing and/or misleading. Monitoring f/VT may be useful when monitoring patient response to changes in the PSV level because it is easily and reliably obtained, varies directly and proportionatelr; with WOB,,, and is more familiar and meaningful to clinicians caring for mechanically ventilated patients. REFERENCES 1. Esteban A%,Frutow F, Tobin MJ, Ma 1, SolsonaJF, S’alverdu I. et al. 4 comparison of four methods of weaning patients from mechanical ventilation. N Engl J Med 1995;332:345-50 2. MacIngTe NR. Weaning from mechanical ventilatory support: volume-assisting intermittent breath versus pressureassisting every breath. Respir Care 1988;33:121-5. 3. Yang IX. Tobin MJ. X prospecti\re study of indexes predicting the outcome of trials of weaning from mechanical ventilation. N Engl J M/led 1991;324z1445-50. 4. Jabour LR, Rabil DM, TruwitJD, Rochester DF. Evaluation of a new weaning mdes baed on ventilatory endurance and the efficiency of gas exchange. Am Rev Resplr Dis 1491;144:531-7 5. Fiastro JF. Habib MP, Shon BY. Campbell SC. Comparison of standard weaning parameters and the mechanically FenMated patients. Chest 1988:9~4.32338. 6. Shikora SA, Benotti PN, Johann&man Jh. The oxygen cost of breathing may predict weaning from mechanical \;entilation better than the respiratory rate to tidal volume ratio. Arch Surg 1994;129:269-74. 7. Krieger BP, Ershowsky PF, Becker DA, Gazeroglu HB. Ewluzluation of conventional criteria for predicting successful weaning from mechanical venttlatory support in elderly patients Grit Care RIed 1989,17:85&61. 8. Lee KH, Hui w, Chan TB, Tan 1\‘C, Lim TK. Rapid shallow breathing (frequencv-tidal volume ratio) did not predict extubation outcome. Chest 1994:105:540-3. 9. Shtkora SA. Bistrian OR, Borlase BC, et al. Work of breathing: reliable predictor of weaning and extubation. Crit Care

Szcrgq October 1997 Med 1990;18:157-62. 10. Gluck EH, Barkoviak MJ. Balk RA, Casey LC, Silver MR. Bone RC. Medical effectiveness of esophageal balloon pressure manometry in weaning panents from mechanical ventilatton. G-it Care bled 1995;23:50-1-9. 11. Levy IV, Miasaki A, Langston D. Work of breathing as a weaning parameter in mechanically ventilated patients. Chest 1995;108:101&20. 12. Baydur A. Behrahs P, Zin WA, et al. X simple method for assessing the validin; of the esophageal balloon technique. Am Rev Respir Dis 1982;126:788-91. in 13. Otis AB, Fenn 14’0, Rahn H. The mechamcs of breathing man. J Appl Physiol 1950;2:592-607. 14 McIlroy MB, Marshal R, Christie RJ. The work of breathing in normal subjects. Clin Sci 1954;13:127-34. 15. Ballantme TVN, Proctor HJ, Broussard NP, et al. The work of breathing: potential for climcal application and the results of studies performed on 100 normal males. Ann Surg 1970:171:5904. 16 Proctor HJ, Woolson R. Prediction of respiratory muscle fatigue by measurements of the work of breathing. 1973;136:367-72. 17. Peters R. Work of breathmg and abnormal mechanics. Med Clin North Am 1971;54:95566. 18. Otis AB. The work of breathing. In: Fenn WO, Rahn H, editors. Handbook of physiologv respiration, Section, Volume 1. Washington, DC: American Physiological Socieq 1964. p

463-78. 19. Sassoon CSH, Light RW, Lodia R, Sieck GC, hlahutte CK Pressure-time product durin g continuous positive airwa! pressure, pressure support ventilation, and T-piece during weaning from mechanical ventilation. Am Rev Respir Dis 1991;143:469-75. CB; Blanch PB. 20 Banner MJ; Kirby RR. Kit-ton OC, DeHaven Breathing frequency and pattern are poor predictors of work of breathing in patients receiving pressure support ventilation. Chest 1995;108.1338-44.

DISCUSSION Dr. Sidney F. Miller (Dayton, Ohio). The breathing frequency,/tidai volume ratio has been advocated by some to predict weaning failure in the clinical setting. Others, however, have not been able to confirm its usefulness, and many have suggested monitoring the work of breathing during the weaning process. Since the work of breathing appears to increase over time, is the SO-minute changes? Correlations

equilibration were

made

period at 5,1O,

long

enough

to detect

15, and 20 cm of pressure

suppo~ and there is an inverse relationship between both work of breathing

and

the i?equency/tidal

volume

ratio

level of pressure support. It seems, therefore, that indicators are best for confirming the need for support. Your data suggest, however, that there gence of the correlation (work of breathing and cy/tidal volume ratio) at lower levels of pressure Can you conclude that either of these values is

with

the

these two pressure is a diverfi-equen-

support. helpful

in

determining the lack of need for pressure support? Statistically, do the two still correlate at these lower levels? In Table II, labeled “unacceptable work” but referred to in your presentation as “acceptable work” at pressure sup port levels of 5 and 10, the work of breathing identified

Johannipan

53% and 67% of patients displaying “unacceptable work,” whereas the frequency/tidal volume ratio identified 73% and 50%. Please clari@ these differences. Finally, 60% of the study patients required tracheostomy before extubation. Were work of breathing measurements or the frequency/ tidal volume ratio helpful in identifying earlier the patient who needed a tracheotomy? Dr. Johannigman. In response to your first question, was the 30-minute period of time adequate for stabilization, anecdotally, although we were just reviewing the data and did not quantitate it, our obsemation was that the stabilization of these parameters occurred about 15 minutes after the adjustment. These were patients who were chronically ventilated and who were going through a long, slow weaning process. You also make the interesting observation regarding the relationship of pressure support and these two parameters and the skew that occurs during T-piece trials. The skew is greatest with the pressure time product, but it also aEect.s the work of breathing to a lesser extent. As patients are withdrawn from pressure support, a greater disparity is observed. I believe this is because each of rhese parameters reflects the fact that there are mam diierent ways to quantitate the work of breathing. One of these ways is the mechanical work of breathing, but this does not account for all the work a patient may be doing. For instance, if a patient is breathing against an occluded endotracheal tube, the intercostal muscles are firing, there is work, and there is energy being consumed. Yet if you use a mechanical definition that requires volume to be moved, then some would say no work is being done. That is why pressure time product and work of breathing vary. Pressure time product is able to include isovolumetric work (e.g.. to trigger the ventilator). The unique characteristic of the rapid-shallow breathing index is that it includes tidal volume as a measure with the parameters. As pressure support increases, tidal volume increases. There should be some expected disparity, because one of these involves the parameter that is being controlled during a trial of pressure support. Increasing pressure support increases tidal volume and thus changes one of the parameters of the rapidshallow breathing index directly. How we account for that at the time of T-piece trial is difficult to explain. As you pointed out, the disparity is still the biggest prob lem. Thhich one of these parameters do you use and what, if anything, is a gold standard? The answer is that no single parameter works well. Work of breathing predicted that approximately 40% to 50% of patients at a level of 5 to 10 cm of pressure support weren’t ready for extubation, whereas the rapid-shallow breathing indes had a higher percentage. This observation relates to your final question about tracheostomy: Did the rapid-shallow breathing index predict who needed tracheostomy and who did not? It did not. Other issues such as mental status, tenacity of secretions, and pro tection of the ainvay must be considered. The patient is often

et al. 741

ready to be extubated from a work of breathing standpoint, but simply doesn’t maintain a clear airway. In these cases, a tracheostomy is needed. Dr. Palmer Q. (Joe) Bessey (Rochester, N.Y). I’ou have shown that the rapid-shallow breathing index and the work of breathing correlate fairly well when you look at the responses of patients to relatively acute changes in ventilato ry support. The other side of weaning, however, is what hap pens in the longer term at 4 hours, 6 hours, and beyond. This relates to fatigue. How do these hue parameters correlate over time, especially as the patient fatigues? The work of breathing would diminish, but what about the rapidshallow breathing index? It might increase or stay the same. How does it relate to work of breathing, fatigue, and long-term weaning success? Dr. Johann&man. There has always been a dichotomy between those of us who are resters and those of us who wean continuously-. The resters take the patient off the ventilator for a couple of hours and then put them back on the ventilator during the evening hours to rest the patient. I prefer to wean patients slowly. increasing their tolerance and endurance with time. I believe the respiratory muscles are endurance muscles that respond more appropriately to distance running than sprinting. Which of these techniques is preferable is still undetermined. A long-term rime prospective study of these parameters is needed. We looked at them only for a short 40 consecutive breath period at the end of a 30-minute trial. If I mayjust comment very briefly about the Bicore. 1 do not find it to be a tool that allows you to wean more rapidly, but it is a valuable teaching tool. I rarely watch the Bicore monitor on a patient without learning something new about the way the ventilator and the patient are interacting. I don’t think it demonstrates the ability to wean faster, but it does teach the clinician and, in doing so, may translate to better patient care. Dr. Christopher I? Johnson (Milwaukee, Wis). I assume that some of these patients had a tracheostomy. It seems intuitive that the work of breathing would be greater through an endotracheal tube than through a tracheostomy. Would you interpret these parameters differently for patients with a tracheostomy versus endotracheal tube? Dr. Johann&man. At the time of the study, none of the patients had undergone tracheotomy, although 9 of the 15 eventually did get a tracheostomy. The work of breathing involved and differences between endotracheal tube and tracheostomy have been described by our group (Davis K, et al. A comparison of the imposed work of breathing with endotracheal and tracheostomy tubes in a lung model. Respir Care 1994). There are differences, but they are not as significant as one might think. You can titrate pressure support based on the size of the endotracheal tube to overcome the imposed work of breathing. In adults, it requires between 7 and 15 cm H,O, depending on the size of the endotracheal tube.