The Use of Expiratory Forced Flows for Determining Response to Bronchodilator Therapy

The Use of Expiratory Forced Flows for Determining Response to Bronchodilator Therapy Samuel F. Boushu, M.D."

Two hundred seventeen patients were studied before and after administration of isoproterenol by IPPB. In 35 patients, lung specimens were examined within 12 months after the study. Patients with elevated bronchial glaud-brouchlal wall ratio (Reid index showed significantly more improvement in forced expiratory flows after bronchodilator therapy, as compared to patients with normal Reid index. There was no difference in response to therapy in patients with emphysema of less than 20 percent, as compared to patients with emphysema of more than 20 percent. Patients with severe reduction in lung recoil pressure had lesser improvement after therapy in forced expiratory flows, as compared to patients with moderate reduction in recoil pressure or normal recoil pressure. In 39 patients inspiratory pulmonary resistance and forced expiratory flows were measured before and after therapy. In all these 39 patients, airway resistance showed comparable improvement regardless of the value for recoil pressure. The lack of response in expiratory flows is attributed to the increased collapsibility of the large airways and not to lack of response to therapy.

I soproterenol is widely used as a bronchodilator in

treatment of patients with chronic obstructive pulmonary disease (COPD). Changes in flow rates after administration of bronchodilators have been advocated as a test to determine the "reversibility" or "irreversibility" of the disease process,' to predict which patients will receive benefit from nebulized bronchodilators," and in some reports" lack of response to isoproterenol has been used as a criterion for a diagnosis of emphysema. Others' have reported patients with COPD and emphysema (judged on roentgenographic evaluation) as having no changes in forced expired volume in one second (FEV l ) after bronchodilator therapy, but significant improvement in inspiratory flow rate. Curtis ct aP studied the response to bronchodilators in patients with COPD over several years, and found almost all will show improvement in FEV 1 after use of bronchodilators on one or more occasions. Some patients may show improvement on one study and no change on another." A recent study? re_._---_._------

"Chief, Pulmonary Function Laboratory, Veterans Adminisstrution Hospital, and Assistant Professor of Medicine, Baylor College of .\1eelieine, Houston. Reprint requests: Dr. Boushi], VA Hospital, 2002 Holcombe Blvd., Ilonston 77031

534

ported patients responding to bronchodilators had elevated airway resistance, and those with normal airway resistance had no change in How rates after such therapy. In this report 217 patients were studied before and after administration of isoproterenol through intermittent positive pressure breathing machine (lPPB). In 35 patients lung specimens were available within 12 months after the study, and the response to treatment with bronchodilators was correlated with morphologic emphysema and bronchial gland hyperplasia. Lung recoil pressure and the single breath method diffusing capacity for CO were utilized to divide the patients into those with severe emphysema (.50 percent or more), and those without emphysema or with mild emphysema (20 percent or less ),8 and the two groups were compared to ascertain the differences in response to isoproterenol inhalation. MATERIALS AND METHODS

All patients in this study had FEV I below 80 percent normal. They were referred to the pulmonary function laboratory for evaluation of lung disease, or were a part of a longterm followup study of COPD. The patients were examined and a questionnaire pertaining to the respiratory tract was

CHEST, VOL. 62, NO.5, NOVEMBER, 1972

535

FORCED FLOWS IN DETERMINING RESPONSE TO BRONCHODILATOR THERAPY

MEFV

/ ¥max 50% VC FIGURE 1. Calculation of midflow rates from maximum expired flow volume curves ( MEFV) and forced vital capacity (FVC) curve in one patient. Upper portion are curves before treatment and lower portion after treatment. Solid lines were not corrected for lung volume, but drawn for each curve separately. Broken lines in the lower curves were drawn at the same volume assuming no change in total lung capacity.

=

--'

= =-

/. FEF 2s.75 -

ADJUSTED

VC

TIME Iseconds I

VOLUME (LI

completed by the examining physician. The study was done in clinically stable patients, and administration of bronchodilators was discontinued ten hours before the test. After completion of bronchodilator therapy, the patient was asked by the technician whether he "improved" or not. During this study only four technicians administered the treatment and did the spirometric studies. All these technicians received adequate training before proceeding on their own. All patients had three forced vital capacity (FVC) curves before the study on a 13.5 L Collins spirometer. Subsequently, they received eight drops of isoproterenol (Isuprel 1:200), diluted by 16 drops of saline, nebulized by IPPB (Bennett PR2). This lasted an average of 10 to 15 minutes. Five minutes after thc IPPB was finished, three FVC curves were repeated. Forced expired volume in one second (FEV\) and maximum mid-expiratory flow rate (FEF2;;-75) were calculated from the record and the highest values used for the study. On another day, and within a period of a few days, 186 patients had single breath diffusing capacity for carbon monoxide and total lung capacity (TLC) by the closed helium method, 1,55 patients had lung recoil pressure at total lung capacity (Pst), and 173 patients inspiratory pulmonary resistance. To correct lung recoil pressure for lung volume, percent normal TLC was divided by the value for recoil pressure at total lung capacity (% N TLC/Pst). Methods of pulmonary function and source of normal values have been described in detail previously.e!" Pleural pressure was measured by esophageal balloon, and inspiratory pulmonary resistance was measured at mid-inspiration. Volume and flow were measured by hi-fi Med Science spirometer. In 25 randomly selected patients airway resistance at 0.5L/sec of flow and lung volumes by constant volume body plcthysmograph.t i recoil pressure at total lung capacity, work of breathing and maximum expiratory flow volumc curves (l\IEFV) were done before and after administration of isoproterenol, in addition to the FVC curve. In an additional 14 patients inspiratory pulmonary resistance was determined before and after use of bronchodilators, in addition tc the sh,dies just mentioned. Work of breathing was calculated from volume-pressure loops recorded on an X-Y recorder. For the ?\IEFV curves, flow and volume from the Ohio Medical spirometer were recorded on a Tektronix storage oscilloscope and photographed. Since flow rates

CHEST, VOL. 62, NO.5, NOVEMBER, 1972

.....

:IE

change with lung volume, midflow from the FVC curve (FEF2".7:;) and from MEFV (V max 50 percent VC) were calculated at the same lung volume, assuming there was no change in total lung capacity (Fig 1). Normal values for TLC and FEV \ were taken from the VA Cooperative Study.I2 Inspiratory pulmonary resistance above 4 em of water per liter per second was considered abnormal.!" In 35 patients, lung specimens were available within 12 months after the study. There were 5 lobectomies, 12 pneumonectomies, and 18 lungs from autopsies. The surgical specimens were from patients with bronchogenic carcinoma, but the tumor masses were small and did not interfere with

RESPONSE TO ISOPROTERENOL

....== .... 0:>

.......

"'"

<>::

~

:z;

= ~

>-

i!i

........ ......., ~

<>::

"'" =

~

0 :>

"'"

160 140 120 100 80 60 40 20 0 20 40 60

..

·• . .. ,·... •. • ••. · •• 1: ·. ·. • · . -------I! #... ,.. ·..+ ·•"..•• .. 11 I.

~

.

: ••

I

FEY 1

FEF

Ymax

25-75

50% VC

I'

...L _____

Gaw/IGV

INSP.

WORK of

RESIST. BREATHING

2. Changes after isoproterenol expressed as percent of pretreatment in 2.5 patients who had all the tests shown. Horizontal lines are the mean for each test. FEV\, f.irced expired volume in one second; FEF 2:;-75 mid-expiratory flow measured from FVC curve and corrected for volume; V max 50 percent VC, mid-expiratory flow measured from flow-volume curves and corrected for volume; Gaw/tGV conductance per liter of thoracic gas volume; inspiratory resistance, inspiratory pulmonary resistance. FH;URE

536

SAMUEL F. BOUSHY

RESPONSE TO ISOPROTERENOL

320 • 300 • 280 • 260 • z:: 240 :E • • -e r• • 220 ..n c:-... , ...:.. ..... • ..... 200 ..... •• •• c=t 180 • ~ • • • • ........ 160 140 .;. -..;: . 120 .

---

r = 0.71

•

LA.I

LA.I

C l..

.

Lon

LA.I

,

.. ... ..,•• .

.;.

:"

tit·

normal, emphysema is severe, and when both are normal there is either no emphysema or emphysema is less than 20 percent of lung parenchyma.

•

110 120 130 140 150 160 170 FEY 1

% OF PRE·TREATMENT

3. Improvement in mid flow rates (FEF 25 - 75 ) and forced expired volume in one second (FEY I ). Increase in FEF 25 75 was slightly above twice that for FEY!; however, .71) and significant the correlation coefficient was high (r (P < .00 I). Patients with decrease in flow rates after treatment were omitted. FIGURE

=

quantitation of emphysema, nor the pulmonary function studies. The lungs were inflated and fixed by formalin steam.l 4 Emphysema was quantitated by the point-counting method, and bronchial gland hyperplasia by the bronchia-l gland-bronchial wall ratio (Reid index). Patients were also divided into those with mild or no emphysema (20 percent or less) and those with severe emphysema (50 percent or more ) on bases of physiologic methods. This separation is possible by the use of the single breath diffusing capacity and recoil pressure at total lung capacity.f 'When both are abo

RESULTS

Changes after use of isoproterenol for various tests in 25 patients are shown in Figure 2. The percent increase in FEV! was less than for mid flows, work of breathing, and airway resistance. Mean improvement in FEV! was 11 percent; FEF 25-75 , 22 percent; V max 50 percent VC, 24 percent; work of breathing, 27 percent; inspiratory resistance, 40 percent; GW ITGV, 16 percent. Seven of 25 patients showed a decrease in conductance. All of these seven patients had evidence of airway collapse even at low flow rates of 0.5L/sec in that the record showed they reached Row plateaus at this Row. In these 25 patients, total lung capacity and recoil pressure showed no significant change from pretreatment value. As expected, vital capacity increased and residual volume and functional residual capacity decreased. For the patients as a whole, percent increase for FEF25_75 (corrected for volume) and FEV! had high correlation (r = .71, Fig 3). The percent increase in FEF25-75 was more than twice that for FEV!, however. For the rest of the results FEF2.5-75 was used to assess the change after isoproterenol inhalation. The volume corrected FEF 2 5 ' 75 will be referred to as isovolume FEF25-7.5. Figure 4 relates the increase in Row rates to pathologic examination of the lung. Patients with emphysema ahove 20 percent had an increase in FEF25 -75 of 15 percent, and those below 20 percent emphysema, FEF 25-75 increased by 19 percent. These differences are not significant. Patients with a normal Reid index had a small increase (8 percent), as compared to patients with elevated Reid index (25 percent) (.05>P>.02).

RESPONSE TO ISOPROTERENOL

4. Changes in the isovolume midflow rates and morphologic emphysema and bronchitis. Horizontal lines are mean values. The difference between patients separated on the basis of Reid index was significant (.05>P>'02). The difference he tween patients separated on the basis of emphysema (portion on left) was not significant ( .70>r>'60). FIGURE

--

160 :z: 140 ..... ...., :E r-e 120 ....,I ..... cc

C'.I

...... .....I 100 ..... ...... cc c-

~

80 60 40

•

~ .1

• •• •• ..,••

: 1

••

--L-

•

,.

•• <20

•• • •• ••

......•

•• ~20

% EMPHYSEMA

• •• •• ••

.........

·• :•

• ••

•

S.44 >.44 REID INDEX

CHEST, VOL. 62, NO.5, NOVEMBER, 1972

537

FORCED FLOWS IN DETERMINING RESPONSE TO BRONCHODILATOR THERAPY

RESPONSE TO ISOPROTEREROL

:z: .... .... ....

-.... 0-

~

U">

r-

.r. .,....

0-

..... ...:. .... ..... ..... 0-

CL.

<::>

~

240 220 200 180 160 140 120 100 80 60 40

J

•

·. 2L ...·. ·. ·. • ••

NONE-MILO

• • • •• s·•:

·.

... ---. 'I"·•

··.

SEVERE

EXTENT Of EMPHYSEMA JUDGED BY DL AND RECOIL PRESSURE .5. Patients with reduced diffusing capacity (less than 3.5 ml/L) and reduced lung recoil pressure (Pst) (%N TLCIPst, more than 6) were con-idered to have severe emphysema (50 percent or more). Patients with normal DL and Pst were considered to have mild or no emphysema (20 percent or less). Horizontal lines are mean values. The difference was statistically significant (P< .001). FIGURE

Figure 5 separates the patients on physiologic bases into those with severe emphysema and mild or no emphysema. Those with severe emphysema had a 25 percent increase in flow rates after administration of isoproterenol, and patients with mild or no emphysema had 44 percent improvement in FEF 25-75 (P < .001). . Further analysis showed this difference to be due to the patients who had severe reduction in Pst (%N TLCjPst greater than 10). Patients with normal recoil pressure (Pst) and moderate reduction in Pst had a similar increase in FEF 2 5 75 after isoproterenol therapy, but in patients with severe reduction in Pst improvement in FEF 25-75 was significantly less (Fig 6) (P < .01). In .39 patients with Pst measurements inspiratory pulmonary resistance was measured before and after treatment in addition to FEF 2 .5 75 . In this group of 39 patients, those with severe reduction in Pst had lesser response in midflow, but equal improvement in inspiratory resistance after treatment, as compared to those with normal or moderately reduced Pst (Fig 7). Thus, the lack of improvement in FEF 2 5 - 75 cannot be attributed to irreversible airway disease in these patients. Patients with elevated airway resistance had a similar increase in forced expiratory flow rates to those with normal airway resistance (Fig 8) . Thirty-six patients were restudied after an inter-

CHEST, VOL. 62, NO.5, NOVEMBER, 1972

val of 12-36 months. The mean increase in FEF2 5 75 on the first study was 36 percent, and on the second study 26 percent. Considering greater than 20 percent increase in FEF2 5 - 75 after use of isoproterenol as representing good response, and below 20 percent as poor response, 2.5 patients responded well on the first study and 11 patients responded poorly. Of the 2.5 patients who showed good response on the first study, 12 showed poor response on the second study, and of the 11 patients with poor response on the first study four showed good response on the second study. The difference between the first and second study was not statistically significant, but this may suggest that response to isoproterenol therapy is diminished later in the course of the disease. To examine this point further, changes in FEF 25 -75 after use of isoproterenol were plotted against FEV! percent of normal. The changes in flow rates, after isoproterenol therapy were not related to thc severity of the disease as judged by FEV! except in patients with minimal evidence of airway obstruction. The increase in FEF 25-75 was identical in patients with

RESPONSE TO ISOPROTERENOL

320 300 280 260 240 ..... :z: ..... 220 :2: ..... 200 ..... co:: I-;180 ..... co:: 160 ~ ...., .....I 140 ...., c.... 120 "'..... "'100 80 60 40 20 00IlC

Q..

•

•

•

•

•

•

• •

• • ••

• •• • :•

.

I •

• ••

" It

1f • •• ·• ,•

• I

-n-

I •

•• •

<6

et·

•

•

,, ·,

I

I

6-9 He % NI Pst

~10

6. Changes in isovolume FEF2o-75 related to recoil pressure measurement. Horizontal lines are mean values. The difference hetween the two groups on the right side of the figure is significant (P<.Ol). FIGURE

538

SAMUEL F. BOUSHY

o

..... «= z: .............."'"«=

......... c:::a .....

«=

3!!

....

--...

20

:z:

-.-

0

0 0

• ••

0

c.:>

""

• •

00

..... "'" ..... .....

= z:

•

0

40

or:: => !a

..... ....,

«= "'" •«= 20 .....

=

FEr 25-75

•

•

0

60

0

o o

100 80

INSP. RESIST.

•

•

0 0

0

•• •

0

• • • • • ...•

00

8

0 0 0 0

0

0 0

• • ••• •

....A-

COO

0 0 0

••

•••

0

r:f3

•• ••

0

• •

0

0

<6

6-9

?IO

TLC % NI Pst. 7. Changes in inspiratory resistance and isovolume FEF2 ,. 75 in 39 patients who had both measurements before and after isoproterenol inhalation related to recoil pressure measurement. Improvement in FEF 2 G- 7 5 was significantly less in patients with severe reduction in recoil pressure, as compared to patients with more normal recoil pressure. Inspiratory pulmonary resistance showed equal improvement in all groups. F'J(;UHE

FEV l of 10 to 60 percent normal and averaged 35 percent. The 21 patients with FEV I of 60 to 80 percent of normal had only 10 percent increase in flow rates after isoproterenol administration. RESPONSE TO ISOPROTERENOL

240 220 200 180 ..... z: ..... .... ::IE 160 ......... .;....... l:li;! 140 zco. 120 ~ "'" ~ 0 80 60 40

...

~

..

.!.

··.·.:

.Il• ··, ·.,.

.: ,.

-: •

~ ·iI I,

ej

~t. '"

1~·

: ..

~

· . ·..

·.

: I

s4 > 4·7 INSP. PULM. RESISTANCE

7 em H,D IL/see. >

FIGURE 8. Changes after isoproterenol were similar in patients with elevated airway resistance (ahove 4 em H 20/L/sec), and tho,'e with normal airway resistance (equal or below 4 em H 20/Usee) (.40>P>'30).

"Wheezing" and "attacks of bronchospasm" are usually associated with reversible airway obstruction. In Figure 9, patients with COPD and wheezing on auscultation of the chest during spontaneous breathing are compared to those without wheezing, and patients with COPD and a history of "asthmatic attacks" to those without such a history. Only patients seen and examined by the author were included in this evaluation to minimize observer variability. A patient was considered to have "asthmatic attacks" if he gave a history of episodic periods of wheezing and shortness of breath. Patients with "wheezing" on auscultation and those with "asthmatic attacks" had a higher increase in FEF 25-75, as compared to patients without these findings. It should be noted that wheezing was frequently present in these patients (128 of 155 had wheezing). Figure 10 shows changes in flow rates which were virtually identical in patients who stated they improved after the treatment, as compared to those who stated they did not (25 and 32 percent). DISCUSSION The degree of change in function studies after administration of isoproterenol can vary, depending on the interval between treatments, duration between treatment and performance of the test," and the test used. 3 , 16, 17 Bronchodilator therapy was stopped ten hours before the study in all patients in this series, and post-treatment tests done 20 to 3:1 minutes after beginning of treatment when maximal response is reported to occur." Maximum midflow rate (FEF 25. 75 ) adjusted for changes in lung volume, was used to assess response to therapy in this series. When compared with maximum midflow rate from flow volume curves, the results are comparable (Fig 2). Bouhuys, et al.!" reported flow rates measured at the same lung volume to be the most sensitive test in assessing the response to bronchodilators when the diseased airways are the small airways, which is the case in COPD.IS Ideally, measurements of flow rates are best done in conjunction with determination of total lung capacity; however, total lung capacity showed no change after use of isoproterenol in 25 patients in this series, and this experience is similar to others."

Response to Bronchodilator Therapy and Emphysema In 35 patients the response to bronchodilators, assessed on one occasion, did not differentiate patients with morphologic emphysema of less than 20 percent from those with more extensive emphysema. CHEST, VOL. 62, NO.5, NOVEMBER, 1972

539

FORCED FLOWS IN DETERMINING RESPONSE TO BRONCHODILATOR THERAPY

RESPONSE TO ISOPROTERENOL

~

~ FIGURE 9. In the group as a whole, patients with history of episodic wheezing and shortness of breath "astlun~ attacks," and patients with wheezing on auscultation of chest during spontaneous breathing, showed greater improvement after treatment as compared to patients without such findings. For both portions of the graph, tile difference was significant (P<.02).

a :::E

~:; .... .... ~ l:::i

=

~

1 •• •• 240 •• • 220 • 200 180 160 140 120 100 I-----l~~------=~----~r__--____..l~:_ 80 •• • 60 40 ' - - - - - -• L - -

RESPONSE TO ISOPROTERENOL

240 220 200 .... 2: 180 . ... 160 ...,1:1I: .... '"":'" ..... ..., 1:1I: ""'A.. 140 ..... .... ..... ..... 120 ~ 100 80 60 40

-~

c::::l

!

--

----:-I

'f

--

-

--

it . -I·

--

---

IMPROVED NO CHANGE PATIENT'S OWN ASSESSMENT OF SYMPTOMS AFTER THERAPY

FIGURE ro. Changes in How rates after isoproterenol were virtually identical in patients who stated they noticed improvement and those who did not (.70>P>.60).

CHEST, VOL. 62, NO.5, NOVEMBER, 1972

WHEEZING

•

•• •

.

•

·• .•

I)

•• • • ••

iT: • ••

PRESENT

It is desirable to have such correlation on a larger number of patients. Recently we have shown that separation of patients into those with severe emphysema (more than 50 percent) from those with mild or no emphysema (20 percent or less) can be made by physiologic tests." Therefore, the patients were divided on the bases of Pst and DL/L into those with severe emphysema (more than .sO percent), and those with no emphysema or mild emphysema (20 percent). The improvement in

.,

• •

U

i

•

·.

:•

It

jj

,.pr

--ll;-

...

• •

-----Al~

A8SENT

PRESENT ABSENT 'ASTHMA' ATTACKS

FEF2fi-7fi was significantly less in patients with physiologic evidence of severe emphysema. This was due to the small change in FEF 25-75 in patients with severe reduction in Pst (%N TLC/Pst ~ 10), while the response in patients with moderate reduction in Pst (%N TLC / Pst 6-9) was similar to patients with normal Pst. Both groups of patients with moderate and severe reduction in Pst should have severe emphysema (more than .sO percent). Furthermore the lack of improvement in How rates in patients with severe reduction in Pst cannot be interpreted as lack of response to therapy in these patients, since inspiratory pulmonary resistance showed equal improvement with use of isoproterenol in them, as compared to patients with more normal Pst (Fig 7). The lack of response in expiratory forced mid How rates must be explained on diflerent bases than lack of response to therapy. The equal pressure point (EPP) concept of Mead et aFo can be used to offer an explanation. In patients with severe reduction in Pst and increased airway resistance, the EPP is located further upstream from the carina, as compared to patients with lesser reduction in Pst. After bronchodilator therapy relaxation of bronchial smooth muscles renders the large airways more collapsible and the EPP moves toward the carina." Thus, after therapy the upstream segment is longer in patients with severe reduction in Pst as compared to patients with more normal Pst. This increases the resistance of this segment and reduces or overcomes the bronchodilation effect on the small airways. Saunders" observed no change in FEV! after bronchodilator treatment in patients with emphysema but increase in inspiratory How rates. Another cause for the little improvement in How rates after

540 isoproterenol could be reduction in recoil pressure after therapy as reported by YlcFadden et al.22 These authors delivered large doses of isoproterenol, which presumably caused muscular relaxation and dilatation of the terminal lung units. 'When they used smaller concentrations of isoproterenol, as the one used in this study, lung recoil pressure did not change. In 25 patients in this series there were no systematic changes in lung recoil at TLG Therefore, in this study, reduction in recoil pressure after use of bronchodilators cannot be implicated as the cause of the minimal change in flow rates observed in some patients. It is clear that measurements of forced expiratory flow rates may increase, decrease, or show no change after bronchodilator therapy, depending on whether bronchodilation of small airways or increased collapsibility of large airways predominate.

Changes in Airway Resistance after Bronchodilator Therapy Inspiratory pulmonary resistance is not affected by changes in collapsibility of the large airways and should be a useful test to measure bronchodilator response. This is especially the case in patients with severe reduction in recoil pressure. A recent report by Bouhuys et aP6 showed no change in airway resistance after administration of bronchodilators, while midflow rates showed significant improvement. Thus, in some patients changes in small airways may not be reflected by measurements of airway resistance alone. For a more adequate evaluation of the effect of bronchodilator treatment, both expiratory mid flow rates and airway resistance should be used, Airway resistance by body plethysmography is measured at low flow rates and usually before flow plateaus occur. Under these conditions airway resistance is comparable to inspiratory pulmonary resistance.P In some patients with severe COPD, flow plateaus occur even at flow rates of 0.5 Llsec, and measurement of airway resistance in these patients can show considerable variability" and may be influenced by increased collapsibility of large airways. In seven patients, specific conductance decreased after bronchodilator therapy (Fig 2). All these patients showed flow plateaus at flow of 0..5 LI sec. In such patients inspiratory pulmonary resistance should be a more reliable test for measuring the response to bronchodilators. It should be mentioned that with body plethysmography, it is possible to measure both inspiratory and expiratory resistance separately.

SAMUEL F. BOUSHY

Clinical Relevance of Determining Response to Bronchodilators Clinically, the response to bronchodilator therapy did not correlate with the patient's own assessment of benefit from the treatment. In general, patients with wheezing and with episodes of bronchospasm can be expected to show more improvement in flow rates after therapy, as compared to patients without such findings. Enormous variability was present between patients, however. In addition, patients who do not show improvement on one occasion may show improvement on another, and according to Curtis et al," almost all patients with COrD will show good response when tested on more than one occasion. Assessed on one occasion, the response to bronchodilator therapy cannot be relied upon to determine the reversibility or irreversibility of COPD, nor to predict the clinical benefit from such therapy. Crompton'" did not find the response to isoproterenol to be helpful in selecting patients likely to benefit from steroid therapy. It is possible that multiple assessments of the rcsponsc to bronchodilator therapy might be of more clinical value. REFERENCES

1 Knowles JH: Chronic obstructive pulmonary disease. Bronchitis and emphysema. Harrison's Principles of Internal Medicine, New York, McGraw-Hill Book Co, 1971, p 1290 2 National Tuberculosis and Respiratory Disease Association: Chronic Obstructive Pulmonary Disease. A Handbook for Physicians. 1965, p 28 3 Gazioglu K, Kaltreider Nl., Hyde RW: Effect of Isoproterenol on gas exchange during air and oxygen breathing in patients with chronic pulmonary disease, Am Rev Resp Dis 104: 188, 1971 4 Saunders KB: Bronchodilators response patterns in patients with chronic airways obstruction: use of peak inspiratory flow rate. Br Med J 2:399, 1967 5 Curtis JK, Liksa AP, Rasmussen HK, et al: The bronchospastic component in patients with chronic bronchitis and emphysema. JAMA 197 :693, 1966 6 Payne CB, Chester EH, Hsi BP: Airway responsiveness in chronic obstructive pulmonary disease. Am J Med 42:554, 1967 7 Stevens PM, Orman BF, Graves G: Contributions of lung recoil and airway resistance to forced expiratory flow limitations. Am Rev Hesp Dis 100:54, 1969 8 Boushy SF, Aboumrad MH, North LB, et al: Lung recoil pressure, airway resistance and forced flow rates related to morphologic emphysema. Am Rev Resp Dis 104 :551, 1971 9 Boushy SF, Kohen R, Billig DM, et al: Bullous emphysema: Clinical, roentgenologic and physiologic study of 49 patients. Chest 54:17,1968 10 Boushy SF, North LB, Trice JA: The bronchial arteries in chornic obstructive pulmonary disease. Am J Med 46:506, 1969 11 DuBois AB, Botehlo SY, Bedell GN, et a\: A rapid plethysmographic method for measuring thoracic gas

CHEST, VOL. 62, NO.5, NOVEMBER, 1972

FORCED FLOWS IN DETERMINING RESPONSE TO BRONCHODILATOR THERAPY

12

13 14 15 16 17 18

volume: a comparison with a nitrogen washout method for me,FlITing Functional residual capacity in normal subjects. J Clin Invest 35:322, 1956 Boren HG, Kory RC, Syner JC: The Veterans Administration-Army Cooperative Study of Pulmonary Function II. The lung volume and its subdivisions in normal men. Am J Med 4U)6, 1966 Frank NR, \lead J, Ferris BG: The mechanical behavior of the lungs in healthy elderly persons. J Clin Invest 36: 1680, 1957 Greenberg SD, Boushy SF, Jenkins DE: Chronic bronchitis and emphysema: Correlation of pathologic findings. Am Rev Resp Dis 96:918, 1\;)67 Mu-hm G J: Time factor in the measurement of response to bronchodilators, Thorax 22:538, 1967 Bouhuys A, Van De Woestinine KP: Respiratory mechanics and dust exposure in Byssinosis. J Clin Invest 49: 106, 1970 Snider GL, Shapiro DA: Patterns of response of spirometric indices to bronchodilator drugs in patients with obstructive lung disease. Am J 1vled Sci 250:367, 1965 Hogg JC, Macklem PT, Thurlheck W1\I: Site and nature of airway obstruction in chronic obstructive lung disease.

541

N Eng J \fed 278: 1355, 1968 19 Afschrift \1, Clements J, Peeters R, et al: Maximal expiratory and inspiratory flows in patients with chronic obstructive pulmonary disease. Am Rev Resp Dis 100:147,19()9 20 Mead J, Turner S1\I, Macklem PT, et al: Significance of the relationship between lung recoil and maximum expiratory flow. J Appl PhysioI22:95, 1967 21 Bouhuys A, Van De Woestinjne KP: Mechanical consequences of airway smooth muscle relaxation. J Appl PhysioI30:()70, 1971. 22 McFadden En, Newton-Howes J, Pride NB: Acute effects of inhaled isoproterenol on the mechanical characteristics of the lungs in normal man. J Clin Invest 49:779, 1970 23 Grimby G, Takioima T, Graham W, et al: Frequency dependence of flow resistance in patients with obstructive lung disease. J Clin Invest 47: 1455, 1968 24 Bouhuys A, Johnson B: Alveolar pressure, airflow rate and lung inflation in man. J Appl PhysioI22:1086, 1967 25 Crompton GK: A comparison of response to bronchodilator drugs in chronic bronchitis and chronic asthma. Thorax 23:46, 1968

Illusion of Depth on the Flat Surface of a Painting The genius who broke the spell of Byzantine conservatism was the Florentine painter Giotto de Bondone (1266-1:3:37). Ciotto's most famous works are wall-paintings or frescoes (so called because they must be painted on the wall while the plaster is still fresh, that is, wet). It is easy to see the similarity of Ciotto's figure of Faith to the works of the Gothic sculptors. Rut it is no statue. It is a painting which gives the illusion of a statue in the round. We see the foreshortening of the arm, the modeling of the face and neck, the deep shadows in the flowing folds of the drapery. Nothing like this had been done for a thousand veal's. Giotto had rediscovered the art of creating the il u~tion of depth OIl a flat surface. For

CHEST, VOL. 62, NO.5, NOVEMBER, 1972

Giotto this discovery was not only a trick to be displayed for its own sake. It enabled him to change the whole conception of painting. Instead of using methods of picture writing he could create the illusion as if the sacred story were happening before our very eyes. Painting, for him, is more than a substitute for the written word. We seem to witness the real event as if enacted on the stage. Giotto was a contemporary of the great Florentine poet Dante, who mentions him in his Divine Comedy. Gombrich EH: The Story of Art (11th ed, revised and enlarged) . London, Phuidon, 1967