Eur Respir J 1989, 2, 543-549

SHORT REVIEW

Flow oscillations on the flow-volume loop: Clinical and physiological implications

W.G. Vincken*, M.G. Cosio*'"

In 1981, SANDERS et al [1) described the "saw-tooth" sign, a sequence of rapid oscillations of the flow-vol­ume loop tracing, as an aid for the detection of sleep apnea in awake patients. Such flow oscillations had been recognised before - in fact ever since the flow­volume loop came into clinical practice at the begin­ning of the sixties - but generally were discarded as a noisy disturbance, created in the measuring instruments or in the subject's airway and blurring the signal con­tained in the flow-volume loop [2-5). To illustrate this, in MILLER and HvArr's important paper [4) on the use­fulness of flow-volume loops in the evaluation of obstructing lesions of the larynx and trachea it can be literally read that the flow plateaux characteristic of upper airway obstruction "... often contained noise or rapid oscillations because of marked flow turbulence ... " and that the figures shown in the article"... show smooth contours for simplicity of display." Smoothing of flow-volume loops has been the preferred approach to display flow-volume loops, not only to meet the need of both manufacturers and users of pulmonary function equipment to get rid of the noisy disturbances, but also to minimize the possible contributions of oscillations to the variability of derived forced expiratory flow rates [3, 6). Some oscillations seen on flow-volume loops may indeed be nothing else but noise. However, recent evidence indicates that flow oscillations in some in­stances, may signal the presence of a functional or structural disorder, usually located in the upper airways.

In this paper we will try to collect the evidence that should change our view on flow oscillations, an ex­ample of how noise may become signal.

Definition

Flow oscillations are defined as a reproducible se­quence of alternating decelerations and accelerations of flow, creating a "saw-tooth" pattern superimposed on the general contour of the flow-volume loop produced by the awake subject (fig. 1). Flow oscillations can occur on any portion of the inspiratory and/or expira­tory parts of the flow-volume loop, whether obtained during forced or tidal breathing.

•Academic Hospital, University of Brussels (AZ VUB), Brussels, Belgium. uRoyal Victoria Hospital, McGill University, Montreal, Canada. Correspondence: Dr.W.Vincken AZ-VUB, Intensive Care Unit, 101 Laarbeeklaan, B-1090, Brussels, Belgium.

Ve

V

VI

VI Fig 1. - Flow-volume loops of 2 patients with the obstructive sleep apnoea synd.rome, showing that flow oscillations may occur during forced eJtpiration as well as during forced inspiration. Volume (V) along the X-axis (each mark is I () and expiratory (~!!) and inspi· ratory ~1) flow along the Y -axis (each mark is 1 l·s).

Contrary to the definition of "saw-toothing" by SAND­ERS et al [1], our definition remains descriptive, pur­posefully avoiding quantitative characteristics such as the amplitude and periodicity (frequency) of individual oscillations, or the number of oscillations in one se­quence. The reason for this is that flow-volume loops are volume- and not time-based recordings, so that os­cillation frequency cannot directly be quantified from them. In order to do so, simultaneously obtained flow­time recordings (fig 2, right panel) or volume integra­tion of the flow-volume loop are required, but these are not always avaiJable in routine pulmonary function test­ing. The number of oscillations in one sequence may vary with the duration of the breathing manoeuvre. In our experience, tidal as well as forced inspiration is shorter than expiration in normal subjects as well as in patients with various pathologic conditions, with the exception of fixed or variable extrathoracic upper air­way obstruction [4]. Therefore an expiratory sequence is likely to contain more oscillatory cycles than an in­spiratory sequence. A minimum of 3 oscillations, as proposed by SANDERS et al. [1) should however be

544 W.G. WINCKEN, M.G. COSIO

present The amplitude of flow oscillations may to a major extent be modified by the damping characteris­tics of the different measuring and recording devices used in different pulmonary function laboratories : underdamped components may overestimate oscillation amplitude, while overdamped system components (such as some flow measuring devices and even some so­called "fast-response" x-y recorders) may underestimate oscillation amplitude.

Clinical correlations

In most papers reporting flow oscillations, a clinical substrate usually has been found to underly flow oscil­lations (table 1).

The "saw-tooth" sign, initially reported by SANDERS et al [1] in patients with the obstructive sleep apnoea syndrome (OSAS), was shown to correspond to rapid fluttering of redundant pharyngeal tissue visible at endo-

.. . ..U:L!l. ~ JJcl:I.!L·I' ., .. .,, . j'i I· t'l trn

" . • , , t I - •• t: ~~ ·' ~<· ,.I :: .. :!;! ~ H ili .:;- .. . , . . .. ~ P ... .. r-· . .. ~ 5Hz ~i·

..... ~ ·

I' ·::1 .•r -l u .. . ,,

.... I o o !.. I

"" ' ' ..... ....... r-I ... . I 1,11 ,! I ....

1:111• ... . ' I'J

I :i .. . . . . ' .! I ; .. I I I~ : :~ 11'

Ve 1

V

.!t 'r .. , ... L.

•• lt ' 'I ..•. .. ··'· ........ I ~ , . .!' J: ; ' ' .. l:l· ..!J! '

... ... _ . ··'· 1 sec . '

I I ,

·-· - • 0 ,.; ·~·· ., .. ! :: . !! , . . - . Tidal . "' '" I-_, • '

breathing I I'

i

0

V time

- rrr . - --- .. --

. . .. _ ...... ..

Vr V

Fig 2. - Row-volume loop (left) and flow-time tracing (right) during tidal breathing in a patient with Parkinson's disease, showing flow oscillations occurring at a frequency of 5 1-b;. Volume (V) along the X·axis (each mark is 0.51 ) and expiratory (V'e) and inspiratory (Vr) flow along lhe Y-uis (each mark is 0.5 l·s). (Reprinted, by permission of lhe New England Journal of Medicine, [51].

For all these reasons, we prefer to regard flow oscillations in descriptive terms as a configurational flow-volume loop pattern, easily recognisable by visual inspection. If quantitative characteristics are required, they might be obtained from time-based flow record­ings. Based on these, the frequency of flow osciiiations observed in awake subjects in our laboratory widely varies between 4 and 60 Hz. Higher oscillation frequen­cies have been recorded in other laboratories, but only in conditions accompanied by audible sound generation such as snoring in sleeping subjects (40-90 Hz) [7] or forced expiratory wheezes (up to 2.7 kHz) [8, 9].

Although the recent introduction of computers for the acquisition and the recording of flow-volume loops has facilitated the calculation of forced in- and expiratory flow rates, it should be realized that most flow-volume loops digitally generated do not have sufficient flow or volume resolution. As a result, computer-steered flow-volume loop recordings may underestimate the frequency content and flow-oscillations may go unrec­ognized. Real-time x-y records obtained on x-y record­ers or storage oscilloscopes therefore should remain the preferred way of graphically displaying flow-volume loops.

scopy. This finding has been amply confirmed since, by a number of authors [10-16] (table 2). In various of these papers, "saw-toothing", although not very sensitive, was considered highly specific for the OSAS, merely because its incidence in the OSAS was com­pared to that in normal control subjects. However, the high specificity of flow oscillations for the OSAS could not be confirmed when comparing OSA patients to control subjects, mostly snorers, referred for excessive somnolence but not having the OSAS [15). This was further substantiated in a large retrospective survey of 2800 flow-volume loops, revealing flow oscillations in 40 cases, an incidence of 1.4% [17]. Of these 40 cases, only 9 (22.5%) had the OSAS, while in a large part of the remaining 31 cases, rather than the OSAS, an­other structural or functional disorder of the upper air­way or its surrounding musculature was present. More particularly, 8 patients had a structural upper airway lesion, 8 had a neurological disorder potentially involv­ing the upper airway muscles, 10 were labelled COPD because of chronic respiratory symptoms and 5 had mis­cellaneous disorders. Three of the 10 COPD patients were later shown to have concommitant or sole upper airway dysfunction characterized by abnormally lax and

FLOW OSCILLATIONS IN FLOW-VOLUME LOOPS 545

unstable upper airway structures [18]. ln these COPD patients now oscillations led to detection of an upper airway disorder which, in the presence of concomitant peripheral airflow limitation, often goes undetected [19, 20).

Table 1. - Clinical conditions associated with flow os­cillations.

Clinical Conditions

Obstructive sleep apnca syndrome (OSAS) Snorers without the OSAS Upper airway stenosis (structural lesions) Upper airway dysfunction Extrapyramidal disorders Neuromuscular disorders (with bulbar involvement) Bum injury of upper airway Leeuwenhoek's disease Herpes Zoster of abdominal muscles

References

[10-16) (15] (17) [18] [25-271 {24]

[22, 23) [38] [39)

Table 2. - Flow oscillation in obstructive sleep apnoea syndrome.

Source (reference) Sensitivity (%) Specificity (%)

SANDERS, 1981 JAMA [1] 85 100 HAPONIK, 1981 ARRD* [10] 55 92 CliAUOIIARY, 1982 ARRD [11) 56 100 TAMMBUN, 1983 ARRD [12] 68 57 RILEY 1983 SLEEP* [13] 50 80 SHoRE 1984 THORAX** [14) 41 92 KRIBGER 1985 CHEST [15] 61 54

-- - - ----------Legend: •, saw-toothing and increased FEFjJFIFj

0 combined;

••. supine flow-volume loop.

It can this be concluded that llow oscillations are not specific for the OSAS, as initially thought, but rather signal the presence of an npper airway disorder (among whjch the OSAS) [17, 21). This has sjnce been sub­stantiated in a number of prospective clinical studies in which now oscillations were recognized to reflect the abnormal state or functioning of the upper airway in burn vjclims with structural upper airway damage [22, 23], in neuromuscular disorders with bulbar (upper air­way) muscle weakness [24] and in extrapyramidal dis­orders with endoscopically visible tremor of tJ1e vocal cords and supraglottic structures [25- 27] (t.ahle 1).

Pathogenetic mechanisms of flow oscillations

Since flow is the result of a driving pressure acting across a flow resistance, flow oscillations can on theo­retical grounds be predicted to result from rapid inter­mittent changes in either driving pressure or airway resistance (table 3).

Rapid intermittent changes in airway resistance can result from several mechanisms. They can he due to ex­cessive turbulence and the Bemouilli effect created past

a structural narrowing of the upper airway, i.e. the "noise" referred to by MiLLER and HYA"n" r4J, or to air­way instability. Airway instability may result from al­tered intrinsic mechanical properties of the airway wall or from loss of the stabilizing and airway patency pro­moting function of the striated musculature surround­ing the upper airway [28-32]. As an example, besides reduced maximal exprratory flow rates, flow oscillations were noticed after instillation of proteolytic enzymes in canine tracheas, which resulted in increased compliance of central airways [33]. Similarly, the incidence of ab­normal flow-volume loop contours suggestive of upper airway dysfunction, such as now plateaus with or with­out flow oscillations, is significantly higher in neuromuscular disorders affecting the bulbar (upper airway) musculature than in neuromuscular disorders without bulbar muscle involvement [24]. ln all t.bese in­stances, flow oscillations probably represent airway walls flutter and flow turbulence leading to small high­frequency (more than 20 Hz) now transients exceeding the maximal flow-volume loop envelope.

Table 3. - Pathogenetic mechanisms of flow oscillations

1. Intermittent changes in airway resistance - structural upper airway lesions

. dynamic upper airway compression (Bemouilli phenomenon) . fluttering of redundant tissue

- upper airway instability . altered mechanical properties of upper airway wall~ . upper airway mu.~cle malfunction

- phasic activity of upper airway muscles 2. Intcrmiucnl changes in driving pressure

- phasic activi ty of respiratory pump muscles. 3. Artefact

- artefact of instrumental origin - lack of patient cooperation (honk, cough. glottic closure)

Another mechanism underlying intermittently chang­ing airway resistance is abnormal phasic activity of the upper airway muscles, leading to visible intermittent airway narrowing and airflow reduction. This has been clearly shown to occur in various extrapyramidal dis­orders, such as idiopathic Parkinson's disease, essential tremor and the Shy-Drager syndrome [25] . In 18 of 27 such patients, flow-volume loops obtained during forced as well as tidal breathing revealed flow oscillations (fig 2) occurring at a frequency of 4 to 8 Hz, which is simi­lar to the frequency of their extremity tremor. These flow oscillations were not accompanied by synchronous pleural pressure oscillations, and hence are not due to tremor of the respiratory pump muscles - the diaphragm noticeably is spared in extrapyramidal disorders [25, 34- 36] - but corresponded to upper airway muscle tremor, clearly visible at endoscopy as regular alternat­ing ab- and adduction of glottic and supraglottic structures at a frequency of 4 to 8 Hz. Since these ab­nonnal movements persisted during breathholding, they do not represent passive fluttering of upper airway structures in the airstream but rather correspond to pha-

546 W.G. YJNCKEN, M.G. COSIO

sic bursts of abnonnal upper airway muscle activity [25].

The second major mechanism that could lead to flow oscillations consists of rapid intermittent changes in driving pressure which could be due to abnormal pha­sic activity of the respiratory pump muscles. This is a very rare occurrence, however, since even in Leeuwenhoek's disease, characterized by rhythmic myoclonus of the diaphragm, the uvula-palatal and la­ryngeal muscles [37), flow oscillations, when reported, where shown to be due to myoclonus of the upper air­way rather than respiratory pump muscles [38].

The only hitherto reported instance of abnonnal res­piratory pump muscle activity underlying the flow oscillations consists of a patient with Herpes Zoster involving the abdominal muscles [39]. In this patient phasic bursts of abnonnal abdominal muscle activity, demonstrable by electromyography, were coupled with mechanical activity producing abnonnal motion of the abdominal wall (detected by inductance plethysmogra­phy) and distinct gastric and oesophageal pressure os­cillations synchronous to the flow oscillations. This is in contrast to the findings in extrapyramidal disorders where oesophageal pressure oscillations were noticea­bly absent during the production of flow oscillations [25).

Finally, before attributing a pathological significance to the observed flow oscillations, artefactual or spuri­ous flow oscillations, such as those created by exces­sive turbulence in the instrument's tubing or by lack of patient co-operation, should be excluded. Non-sustained oscillations lacking reproducibility on different days or with different instrumental set-up, are more likely to represent artefacts. Particular attention should be given to avoid either sound production ("honks", or high- fre­quency oscillations which usually occur around peak flow) or cough and voluntary intermittent approxima­Lion and closure of the vocal cords (low-frequency, large excursions of the now tracing between the maxi­mal flow-volume loop envelope and the volume axis, which usually occur at low lung volume). Figure 3) il­lustrates these 2 types of spurious flow oscillations voluntarily generated by the subject.

In general, flow oscillations are in most - if not all - instances the result of intermittent changes in air­way resistance. In only a rare instance, phasic activity of the respiratory pump muscles may cause flow oscil­lations, but then synchronous oesophageal or gastric pressure oscillations, should be demonstrable. Further­more, superimposition of flow oscillations on the gen­eral flow-volume loop contour suggests that the rapid intennittent changes in airway resistance take place in the common airway proximal to the carina or not far­ther out than the lobar bronchi. Indeed, even if osci lla­tory flow were to originate in more peripheral airways, it is likely to be damped out by non-oscillatory flow from the many other airway branches, resulting in a smooth flow-volume loop contour recorded at the mouth. Also, oscillatory flow is unlikely to occur more peripherally than the lobar bronchi where linear flow speed decreases due to an increasing cumulated cross-

Ve

10

s

s

10 Vr

Honk

Honk

Glottic closure

V

Fig 3. - Flow-volume loop recorded while the subject was asked to produce sound during inspiration and immediately after the start of forced expiration (producing a honk, i.e., spurious high-frequency oscillations around peak flow) and to close inlermittently his vocal cords during the second half of forced expiration (producing large amplitude oscillations or mini-coughs in the second half of the MEFV curve). Volume (V) along the X-uis (each mark is 1 I) and expiratory ('VE) and inspiratory ('Vr) flow along the Y -axis (each mark is 1 l·s).

sectional area and the flow regime becomes progres­sively laminar.

It is easy to understand how the discussed mecha­nisms can lead to flow oscillations during tidal breath­ing and forced inspiration .. During forced expiration, however, the general understanding is that maximal flow is, at least in the second half of the manoeuvre (the so-called effort-independent part), determined by a complex interaction between elastic lung recoil, airway geometry (the frictional resistance of upstream periph­eral airways) and dynamic airway transmural pressure­area characteristics (or airway compliance) at choke points where flow speed approximates wave speed [2, 40-46). Events downstream from the flow-limiting choke points do not appear to influence maximal expi­ratory flow. However, flow oscillations have been posi­tively identified during this effort-independent portion of forced expiration. This, then, must indicate that the changes in airway resistance and caliber occurring in the upper airway downstream from choke points are of such a magnitude as to interfere with the normal flow limiting mechanisms. Choke points can then be regarded

FLOW OSCILLATIONS IN FLOW-VOLUME LOOPS 547

to jump back and forth between, on the one hand, the upper airway site where the structural or functional disorder resides, and on the olher hand, its more peripheral location appropriate for Lhe lung volume pre­vailing at that time. Similar mechanisms have been proposed to explain the highly reproducible plateau-knee configuration or concave bump which has been noticed to occur following peak flow in some nonnal subjects [47]. These knees, or sudden flow decelerations com­parable to the downstroke of a flow oscillation, were attributed to sudden, more peripheral relocations of air­way choke points [47]. Furthennore, the location and motion of airway choke points during forced expiration were found to be a ltered by small changes in local air­way stresses such as those induced by changes in body posture [48] or neck position [49].

Funct ional implications of flow oscillations

Flow oscillations related to the presence of an upper airway disorder are often accompanied by physiologi­cal evidence of upper airway obstruction (UAO). The lauer can be inferred from an excessive response to breathing of a low-dens ity gas mixture, such as hcliox (20% oxygen in helium) (20), or from an incre.ase in the following indices: the ratio of the fo rced expired volume in 1 s (PEV1) to the forced expired volume in 0.5 s (FEV/FEV 0_5) , tile ratio of the FEV

1 to Lhe peak

expiratory flow (FE V t/PEF), tbe ratio of the forced mid­expiratory to the forced mid- ins pirato ry flow (FEF5JFIF5~ or the peak inspiratory flow (PrF) l50].

Thus, in a retrospective review of 2800 now-volume loops, physiologic evidence of UAO was present in 14 (35%) of the 40 instances with flow osciiJations, inc lud­ing 4 of 8 patients with a structural upper airway le­sion and 10 of 32 patientc; without such a lesion [17). Similarly, of 27 patients with an extrapyramidal disor­der, physiologic evide nce of UAO was present in 10 (37%) patients, 9 of whom presented flow oscillations on !.he ir flow-volume loops [25]. In bum victims, flow­volume loop abnormalities including flow oscillations and physiologic evidence of UAO were shown to cor­relate with structural upper airway changes observed during nasopharyngoscopy and were shown to be useful in Lhc early assessment of patients at risk for UAO and the eventual need for endotracheal intubation [22, 23].

The detection of flow oscillations on flow-volume loop should instigate additional examinations directed at tile upper airway in Lhe search for a structural or func­tional abnormality with or without physiologic evidence of UAO. The observation of flow osciiJations in a struc­tural upper airway lesion witlloU£ physiologic evidence of UAO suggests !.hat fl ow oscillations arc more sensi­tive to structural upper airway lesions than the conven­tional spirometric criteria of UAO. Probably, then, the structural lesion is not severe enough to narrow Lhe air­way lwnen down to a diameter of less tllan 8 mm , tile critical dimension below which spirometric evidence of UAO appears (4]. but is severe enough to create ex-

cessive turbulence or a Bemouilli effect on the airway walls downstream from the lesion, resul ting in flow oscillations.

If a structural lesion cannot be found, a functional disorder of the upper airway, due to altered mechani­cal properties of its walls or to malfunction of !he up­per airway muscles, that in normal conditions stabilize and maintain patency of the upper airway, s hould be suspected.

Such a functional disorder may or may not be accom­panied by physiologic evidence of UAO, but in the lat­ter event tile presence of Oow oscil lations should be regarded as a sensitive sign, premonitory for an ongo­ing upper airway disorder, not yet of consequence but eventually leading to signi ficant UAO.

Conclusion

With proper attention g iven to the instrumental set­up and to the correct execution and acquisition of the flow~volume loop test. flow oscillations represent a readily identifiable sign with potentially valuable diag­nostic implications. Although a possible mechanism underlying the now oscillations may be rapidly inLer­mittent phasie activity of the respiratory pump muscles, in most ins tances flow oscillations represent rapid in­terrnillent changes in airway resistance and calibre, due to a structural or func tional upper airway disorder. The detection of flow oscillations on now-volume loops, should thus instigate extensive investigations primarily directed at the upper airway and its surrounding mus­culature. In recent years, the importance of the func­tional integrity of the upper airway, including iLS surrounding musculature, has been much e mphasized. This "organ" regulates the flow of air to and from the lungs in a way integrated wilh Lbe remainder of Lhe res­piratory system. Its function is often difficult to assess and, if not specifically looked for, malfunction may go undetected. Attention to the now-volume loop configu­ration may help to assess tlte functional status of the upper airway. Abnormal contours, such as flow oscil­lations, appear to be sensitive premonitory markers of ongoing djsorders with porenliaiJy serious consequences. Simple visual inspection of the now-volume loop con­lOur should become an integral part of pulmonary func­tion testing.

Acknowledgements: The authors thank Mrs. Cannon Dan1uay and her staff of lhc Desmond N Stoker Pubnonary Function Laboratory of the Royal Victoria Hospital , Me Gil l University, Montreal, Canada. The authors also lhank Mrs. Hilda De Backer for expert secretarial assistance.

Refer ences

I. Sanders MH, Martin RJ, PeMock BE. Rogers RM. - Tite detection of sleep apnea in lhe awake patient. The "saw­toolh" sign. lAMA. 1981, 245, 2414-2418. 2. Hyau RE, Schildcr DP, Fry DL. - Rclatjonship between ma;dmal expiratory now and degree of lung inflation. J Appl Physiol, 1958, 13. 331-336.

548 W.G. VJNCKEN, M.G. COSIO

3. Clement J, van de Woestijne KP. - Variability of maxi­mum expiratory flow-volume curves and effort independency. J Appl Physiol, 1971, 31, 55-62. 4. Miller RD, Hyatt RE. - Evaluation of obstructing lesions of the trachea and larynx by flow-volume loops. Am Rev Respir Dis, 1973, 108, 475-481. 5. Ycmault JC, Englert M, Sergysels R, De Coster A. -Upper airways stenosis: a physiologic study. Am Rev Respir Dis, 1973, 108, 996-1000. 6. Green M, Mead J, Turner JM.- Variability of maximum expiratory flow-volume curves. J Appl Physiol, 1974, 37, 67-74. 7. Liistro C, Stanescu D, Veritcr C, Rodcnstein DO, Aubert­Tulkens G. - Flow limitations in snorers during sleep. Eur Respir J, 1988, 1, 159s. 8. Gavriely N, Kelly KB, Grotberg JB, Loring SH. - Forced expiratory wheezes are a manifestation of airway flow limitation. J Appl Physiol, 1987, 62, 2398- 2403. 9. Webster PM, Sawatzky RP, Hoffstein V, Leblanc R, Hinchey MJ, Sullivan PA. -Wall motion in expiratory flow limitation: choke and flutter. J App/ Physiol, 1985, 59, 1304-1312. 10. Haponik EF, Bleccker ER, Alien RP, Smith PL, Kaplan J. - Abnormal inspiratory flow-volume curves in patients with sleep-disordered breathing. Am Rev Respir Dis, 1981, 124, 571-574. 11. Chaudhary BA, Elguindi AS, Giacomini K, Speir W A. - Diagnosis of sleep apnea by flow-volume loops. Am Rev Respir Dis, 1982, 125, 101. 12. Tammelin BR, Wilson AF, de Berry Borowiecki B, Sassin JF. - Flow-volume curves reflect pharyngeal airway abnormalities in sleep apnea syndrome. Am Rev Respir Dis, 1983, 128, 712-715. 13. Riley R, Guilleminault C, Herran J, Powell N. - Cepha­lometric analyses and flow-volume loops in obstructive sleep apnea patients. Sleep, 1983, 6, 303-311. 14. Shore ET, Millman RP. - Abnormalities in the flow­volume loop in obstructive sleep apnea sitting and supine. Thorax, 1984, 39, 775-779. 15. Kricger J, Weitzenblum E, Vardevenne A, Stierle JL, Kurtz D. - Flow-volume curve abnormalities and obstructive sleep apnea syndrome. Chest, 1985, 87, 163-167. 16. Pavlakou G, Antoniou J, Rees J. - Features predicting obstructive sleep apnea in patients referred for sleep study. Eur Respir J, 1988, 1, 144s. 17. Vincken W, Cosio MG. - Flow oscillations on the flow­volume loop: a nonspecific indicator of upper airway dysfunction. Bull Eur Physiopathol Respir, 1985, 21, 559-567. 18. Vincken W, Dollfuss RE, Cosio MG. - Upper airway dysfunction detected by respiratory flow oscillations. Eur J Respir Dis, 1986, 68, 50-57. 19. Kryger M, Bode F. Antic R, Anthonisen N. - Diagno­sis of obstruction of the upper and central airway. Am J Med, 1976, 61, 85-93. 20. Lavelle TF, Rotrnan HH, Weg JG. - Isoflow-volume curves in the diagnosis of upper airway obstruction. Am Rev Respir Dis, 1978, 117, 845- 852. 21. Neukirch F, Weitzenblum E, Korobaeff M, Annesi I, Kauffmann F. - Flow-volume curve abnormalities and upper airway obstruction signs in an epidemiological survey. Eur Respir J, 1988, 1, 74s. 22. Haponik EF, Munster AM, Wise RA, Smith PL, Meyers DA, Britt EJ, Bleecker ER. - Upper airway function in bum patients. Correlation of flow-volume curves and nasopharyngoscopy. Am Rev Respir Dis, 1984, 129, 251-257. 23. Haponik EF, Meyers DA, Munster AM, Smith PL, Britt EJ, Wise RA, Bleecker ER. - Acute upper airway injury in

burn patients. Serial changes of flow-volume curves and nasopharyngoscopy. Am Rev Respir Dis, 1987, 135, 360-366. 24. Vincken W, Elleker G, Cosio MG. - Detection of upper airway muscle involvement in neuromuscular disorders using the flow-volume. loop. Chest, 1986, 90, 52-57. 25. Vincken WG, Gauthier SG, Dollfuss RE, Hanson RE, Darauay CM, Cosio MG. - Involvement of upper-airway muscles in extrapyramidal disorders. N Eng J Med, 1984, 311, 438-442. 26. Schiffman PL. - A "saw-tooth" pattern in Parkinson's disease. Chest, 1985, 87, 124-126. 27. Bogaard JM, Hovestadt A, Meerwald J, van de Meche FGA, Stigt J. -Flow-volume curves in Parkinson's Disease. Bull Eur PhysiopaJhol Respir, 1987, 23, 417s. 28. 'Brouillette RT. Thatch BT. - A neuromuscular mecha­nism maintaining extrathoracic airway patency. J Appl Physiol, 1979, 46, 772-779. 29. Strohl KP, Hensley MJ, Hallett M, Saunders NA, Ingram RH Jr. - Activation of upper airway muscles before onset of inspiration in normal humans. J Appl Physiol, 1980, 49, 638-642. 30. Chemiack NS. - Respiratory dysrhythmias during sleep. N Engl J Med, 1981, 305, 325-330. 31. Strohl KP.- Upper airway muscles of respiration. Am Rev Respir Dis, 1981, 124, 211-212. 32. Proctor DF. - All that wheezes ... Am Rev Respir Dis, 1983, 127, 261-262. 33. Jones JG, Fraser RB, Nadel JA. - Effect of changing airway mechanics on maximum expiratory flow. J Appl Physiol, 1915, 38, 1012-1021. 34. Nugent CA, Harris HW, Cohn J, Smith CC, Tyler FH. - Dyspnea as a symptom in Parkinson's syndrome. Am Rev Tuberc, 1958, 78, 682-691. 35. Petit JM, Delhez L. - Activite electrique du diaphragmc dans la maladie de Parkinson. Arch lnJ Physiol, 1961, 69, 413-417. 36. Vincken W, Gauthier S, Dollfuss R, Hanson R, Darauay C, Cosio MG. - Respiratory muscle involvement in Parkinson's disease. N Eng J Med, 1984, 311, 1517. 37. Phillips JR, Eldridge FL. - Respiratory Myoclonus (Leeuwenhoek's disease). N Eng J Med, 1973, 289, 1390-1395. 38. Bollinger SM, Menkes HA, Benjamin JJ, Ball WC Jr. -The syndrome of rhythmic palatal myoclonus. A cause of sig­nificant extrathoracic airway obstruction. Am Rev Respir Dis, 1974, 110, 803-806. 39. Passerini L. Cosio MG, Newman SL. - Respiratory muscle dysfunction after Herpes Zoster. Am Rev Respir Dis, 1985, 132, 1366-1367. 40. Mead J, Turner JM, Macklem PT, Little JB. - Signifi­cance of the relationship between lung recoil and maximum expiratory flow. J Appl Physio/, 1967, 22, 95-108. 41. Pride NB, Permutt S, Riley RL, Bromberger-Bamea B. - Determinants of maximum expiratory flow from the lungs. J Appl Physiol, 1967, 23, 646-662. 42. Dawson SV, Elliott EA. - Wavespeed limitation of ex­piratory flow, a unifying concept. J Appl Physiol, 1977, 43, 498-515. 43. Elliott EA, Dawson SV. - Test of wavespeed theory of flow limitation in elastic tubes. J Appl Physiol, 1977, 43, 516-522. 44. Lambert RK, Wilson TA, Hyatt RE, Rodarte JR. - A computational model for expiratory flow. J Appl Physiol, 1982, 52, 44-56. 45. Pedersen OF, Jngram RH Jr. - Configuration of maxi­mum expiratory flow-volume curve: model experiments with physiological implications. J App/ Physiol, 1985, 58, 1305-1313.

FLOW OSCILLATIONS IN FLOW-VOLUME LOOPS 549

46. Pedersen OF, Ingram RH Jr. - The use of maximum expiratory flow-volume curves on air and He/0~ to assess pe­ripheral pressure losses in the airways. Bull Eur PhysiopaJhol Respir, 1987, 23, 649-662. 47. Tien Y-K, Elliott EA, Mead J. -Variability of the con­figuration of maximal expiratory flow-volume curves. J Appl Physiol, 1979, 46, 565-570. 48. Castile R, Mead J, Jackson A, Wohl ME, Stokes D. -

Effects of posture on flow-volume curve configuration in nom1al humans. 1 Appl Physio/, 1982, 53, 1175-1183. 49. Melissinos CG, Mead J. - Maximum expiratory flow changes induced by longitudinal tension on trachea in nom1al subjects. J Appl Physiol, 1917. 43, 537-544. 50. Rotman HH, Liss HP, Weg JG. - Diagnosis of upper airway obstruction by pulmonary function testing. Chest, 1975, 68. 796-799.