racheobronchomalacia is a condition characterized by excessive T airway collapsibility due to increased

Downloaded from www.ajronline.org by 37.44.207.55 on 01/19/17 from IP address 37.44.207.55. Copyright ARRS. For personal use only; all rights reserved...
Author: Jessie Payne
0 downloads 1 Views 1MB Size
Downloaded from www.ajronline.org by 37.44.207.55 on 01/19/17 from IP address 37.44.207.55. Copyright ARRS. For personal use only; all rights reserved

Baroni et al. Dynamic CT of the Central Airways

Original Report Ronaldo H. Baroni1,2 Simon Ashiku3 Phillip M. Boiselle4 Baroni RH, Ashiku S, Boiselle PM

Dynamic CT Evaluation of the Central Airways in Patients Undergoing Tracheoplasty for Tracheobronchomalacia OBJECTIVE. The objective of this study was to describe the role of pre- and postoperative dynamic CT in patients undergoing tracheoplasty, a novel surgical method for treatment of severely symptomatic tracheobronchomalacia. CONCLUSION. Five patients were referred for dynamic MDCT before and after undergoing tracheoplasty at our institution. Preoperatively, all patients showed signs of tracheobronchomalacia (≥ 50% airway collapse) on bronchoscopy, and four (80%) of these five patients showed evidence of malacia on dynamic forceful expiratory CT scans. In all five cases, postoperative CT showed a reduction in the degree of airway collapse during expiration, changes in shape of the trachea during inspiration, and posterior wall thickening related to the procedure. Our preliminary results suggest a potentially important role for CT in the pre- and postoperative assessments of patients with tracheobronchomalacia referred for tracheoplasty. racheobronchomalacia is a condition characterized by excessive airway collapsibility due to increased flaccidity of the membranous portion of the central airways and weakness of the airway walls and supporting cartilage [1]. The diagnosis of tracheobronchomalacia can be established by the identification of a reduction in the cross-sectional area of the airway greater than or equal to 50% at expiration or during coughing [1–3]. Although bronchoscopy is still considered the gold standard diagnostic tool, CT is gaining increased recognition as a reliable noninvasive method for establishing the diagnosis of tracheobronchomalacia. The treatment of tracheobronchomalacia is variable and depends on the severity of symptoms. In patients with severely symptomatic tracheobronchomalacia due to excessive flaccidity of the posterior membranous portion of the airway wall, surgical intervention can be curative and is considered the treatment of

T

Received January 24, 2004; accepted after revision August 10, 2004. 1Instituto de Radiologia, Hospital das Clinicas da

Faculdade de Medicina da USP, São Paulo, Brazil.

2Instituto Israelita de Ensino e Pesquisa

Albert Einstein, Department of Radiology, Albert Einstein Hospital, São Paulo, Brazil.

3Department

of Cardiothoracic Surgery, Harvard Medical School, Beth Israel Deaconess Medical Center, Boston, MA 02215.

4Department

of Radiology, Harvard Medical School, Beth Israel Deaconess Medical Center, 330 Brookline Ave., Boston, MA 02215. Address correspondence to P. M. Boiselle.

AJR 2005;184:1444–1449 0361–803X/05/1845–1444 © American Roentgen Ray Society

1444

choice [4]. The surgical procedure is referred to as “tracheoplasty” and consists of surgical plication of the posterior wall of the central airways using a graft sutured to the posterior membranous portion [4]. The goals of the surgery are to reshape the trachea, increase its rigidity, and decrease its collapsibility [4]. To our knowledge, the CT findings in patients undergoing this procedure have not yet been described. We report the pre- and postoperative CT findings in a series of consecutive patients who underwent tracheoplasty at our institution. We also briefly describe the technical aspects of this surgical procedure. Materials and Methods During the period between April 2002 and June 2003, five patients with a confirmed diagnosis of severely symptomatic tracheobronchomalacia based on bronchoscopy findings and pulmonary function test results underwent tracheoplasty at our institution and were imaged before and at least 1 month after the surgical procedure. One patient was

AJR:184, May 2005

Dynamic CT of the Central Airways

TABLE 1

Demographic and CT Data for Five Patients with Confirmed Tracheobronchomalacia Before and After Tracheoplasty

Patient

Downloaded from www.ajronline.org by 37.44.207.55 on 01/19/17 from IP address 37.44.207.55. Copyright ARRS. For personal use only; all rights reserved

No. 1

Percentage of Airway Collapse Obtained During Dynamic Expiration Time of CT

Age Sex (yr) 62

Trachea

M Presurgical Postsurgical

2

56

M Presurgical Postsurgical

3 4 5

78 58 60

M Presurgical F

Carina

Right Mainstem Bronchus

Left Mainstem Bronchus

Shape of Trachea During Inspiration

During Expiration

Posterior Wall Thickening

95.0

74.7

65.4

87.7

Biconvex

Lunate

No

40.4

37.5

42.5

75.8

Horseshoe

Horseshoe

Yesa

48.4

51.7

62.6

60.3

Round

Lunate

No

13.9

20.3

25.6

0

Round

Horseshoe

Yesa

41.5

57.0

33.3

74.9

Biconvex

Lunate

No

Postsurgical

4.8

5.1

4.3

1.8

Horseshoe

Horseshoe

Yesb

Presurgical

19.1

27.0

30.4

26.6

Round

Horseshoe

No

Postsurgical

0

0

7.0

7.3

Round

Horseshoe

Yesb

M Presurgical Postsurgical

84.3

85.6

74.0

79.7

Biconvex

Lunate

No

58.7

61.3

70.3

60.4

Horseshoe

Lunate

Yesb

a6 mm. b5 mm.

also imaged 4 days after the operation for a clinically suspected surgical complication. All patients underwent the same CT protocol for both pre- and postsurgical examinations on an 8MDCT scanner (LightSpeed, GE Healthcare) with a gantry rotation time of 0.5 sec. The CT protocol focused on the central airways—from the thoracic inlet to approximately 2 cm below the carina—and included imaging during two phases of respiration: end-inspiratory and dynamic expiratory phases. The end-inspiratory helical scan (170 mAs, 120 kVp, 2.5-mm collimation, high-speed mode, and pitch equivalent of 1.5) was obtained first in all patients and was followed by a dynamic expiratory low-dose helical scan (40 mAs, 120 kVp, 2.5-mm collimation, high-speed mode, and pitch equivalent of 1.5) in which the patient was instructed to forcefully exhale during CT acquisition. The dynamic expiratory component of the CT scan was completed in approximately 5 sec. Our hospital institutional review board approved the review of radiologic and clinical data for this study. Informed consent was not required for this retrospective analysis, but patient confidentiality was protected. All CT scans (pre- and postoperative inspiratory and expiratory images) were jointly reviewed by consensus by two experienced thoracic radiologists on a PACS workstation (Centricity version 2.0, GE Healthcare) for several parameters. The first parameter was quantification of the percentage of airway luminal collapse during expiration for the levels of the trachea, carina, and each mainstem bronchus; this was performed by using an analysis tool available on our PACS software to measure the cross-sectional area of the airway lu-

AJR:184, May 2005

men in millimeters squared on end-inspiratory and dynamic expiratory scans. We calculated the percentage of luminal collapse by dividing the dynamic expiratory cross-sectional area by the endinspiratory cross-sectional area and multiplying by 100. The second parameter was the shape of the trachea. The reviewers noted whether the trachea was round, oval, horseshoe, biconvex, or lunate given that round, oval, and horseshoe shapes are considered normal tracheal configurations [5, 6]. The third and fourth parameters were presence or absence of tracheal wall thickening (defined as wall thickness > 3 mm [7]) and the presence or absence of associated lung or mediastinal disease. With regard to measurement of the airway lumen, cross-sectional area measurements were performed by an experienced radiologist using an electronic tracing tool, and individual measurements were agreed on by consensus review by a second radiologist. Our methodology did not involve averaging of slices because we used only a single axial CT image with the greatest airway collapse at dynamic expiration for each level of analysis (trachea, carina, and mainstem bronchi) and compared it with the respective axial image level at end-inspiration. Postoperative scans were also assessed for specific complications, including airway dehiscence, mediastinal hematoma, abscess, and pleural effusion or pneumothorax. The percentage of reduction in the degree of airway collapse, from the preoperative to the postoperative studies, was also calculated for all five patients.

Results Our patient population consisted of five patients (four men, one woman) who under-

went tracheoplasty for the treatment of tracheobronchomalacia and had both pre- and postoperative CT studies. The mean age of the patients was 62 years (range, 56–78 years). Table 1 lists the relevant demographic and CT data of our patient population. All five patients showed evidence of tracheobronchomalacia at bronchoscopy, and four (80%) of five showed evidence of tracheobronchomalacia on preoperative dynamic CT. One patient (patient 4 in Table 1) who has severe asthma was unable to fully cooperate with the expiratory breathing instructions, resulting in a false-negative CT scan. The mean percentages of airway collapse at dynamic expiration on preoperative scans were 57.6%, 59.2%, 53.1%, and 65.8%, respectively, for the levels of the trachea, carina, right mainstem bronchus, and left mainstem bronchus. For the scans obtained after tracheoplasty, the mean percentages of airway collapse, for the same respective levels, were 23.6%, 24.8%, 29.9%, and 29.1%. With regard to the shape of the trachea on preoperative CT studies, three (60%) of the five patients presented with a biconvex or “fishmouth” tracheal shape at end-inspiration; four (80%) of five patients had a convex anterior–concave posterior or lunate tracheal shape at dynamic expiration. The only patient whose trachea did not exhibit a lunate configuration at dynamic expiration was the patient who was unable to fully cooperate with expiratory breathing instructions (patient 4). Conversely, after tracheoplasty, all five patients showed a normal tracheal shape at end-inspi-

1445

Downloaded from www.ajronline.org by 37.44.207.55 on 01/19/17 from IP address 37.44.207.55. Copyright ARRS. For personal use only; all rights reserved

Baroni et al.

A

B

C

D

Fig. 1.—62-year-old man (patient 1 in Table 1) with tracheobronchomalacia: comparison of pre- and postoperative CT images. A, Preoperative end-inspiratory CT image obtained before surgery shows trachea has biconvex (“fishmouth”) shape at end-inspiration. Incidental note is made of paraseptal emphysema (arrow). B, On preoperative dynamic expiratory CT image, trachea shows marked collapse at dynamic expiration, acquiring lunate shape. C, Postoperative end-inspiratory CT image obtained after tracheoplasty shows trachea has normal horseshoe shape. Emphysema (arrows) is also noted. D, Postoperative dynamic expiratory CT image shows that horseshoe configuration persists despite mild degree of collapse.

ration, and four (80%) of the five patients also showed a normal tracheal shape at expiration (Fig. 1). We observed posterior airway wall thickening (> 3 mm) [7] on CT scans of all five patients after tracheoplasty, an expected finding because the posterior airway wall is folded and reinforced with a graft (Fig. 2). The average thickness of the posterior wall after surgery was 5.4 mm (range, 5–6 mm). In one patient who was imaged 4 days after the operation for a suspected surgical

1446

complication, we observed heterogeneous soft-tissue density surrounding the trachea and esophagus that corresponded to edema and hemorrhage; the surgeon considered these early postoperative findings to be normal. The patient showed a good clinical recovery and later underwent another CT study 45 days after surgery, which showed complete resolution of those early postoperative findings (Fig. 3). In clinical follow-up, all five patients reported subjective improvement of symptoms

after the surgical procedure. None of the five patients experienced a significant complication from the procedure. Discussion Although tracheobronchomalacia often goes undetected, this disorder has recently been increasingly recognized as a relatively common cause of chronic respiratory symptoms in adults [8, 9]. For example, it has been identified in up to 23% of patients with symptoms of chronic bronchitis [8] and in 14% of

AJR:184, May 2005

Downloaded from www.ajronline.org by 37.44.207.55 on 01/19/17 from IP address 37.44.207.55. Copyright ARRS. For personal use only; all rights reserved

Dynamic CT of the Central Airways

A

B

Fig. 2—58-year-old woman (patient 4 in Table 1) with tracheobronchomalacia. A, Unenhanced CT image obtained at end inspiration before tracheoplasty shows normal thickness of posterior airway wall at level just below carina. B, Unenhanced CT image obtained at end inspiration after tracheoplasty shows diffuse posterior wall thickening (arrow). This finding is related to plication of posterior airway wall with reinforcing Marlex mesh.

nonsmoking patients presenting with chronic cough [8]. Moreover, in a recent study, tracheobronchomalacia was unexpectedly diagnosed in 10% of patients undergoing CT pulmonary angiography for suspected pulmonary embolism [10]. The treatment of tracheobronchomalacia is variable, ranging from clinical follow-up for incidentally discovered, asymptomatic cases to various types of intraluminal airway stenting and surgical procedures for symptomatic cases. In the past, aortopexy was the routine procedure for the surgical treatment of this condition [11]. This procedure consists of attachment of the anterior wall of the aorta to the undersurface of the sternum, thus providing sufficient space in the mediastinum so that the malacic tracheal segment would not be compressed by adjacent structures [11]. However, because this procedure does not directly treat the malacic segment, airway collapse may still occur during expiration and coughing. The first article describing a tracheoplasty procedure was written by Herzog and published in 1954 [12]. This procedure has recently received renewed attention for its role in treating symptomatic tracheomalacia in adults [4]. Herzog initially used bone grafts to increase the rigidity of the membranous wall of the trachea [12]. Approximately one decade later, Rainer et al. [13] altered the approach by imbricating the posterior membranous tracheal wall with a polypropylene sheet. More recently, the technique has been further modified

AJR:184, May 2005

by plicating the posterior membranous wall with Marlex mesh [4]. In our institution, a right posterolateral thoracotomy approach is used for tracheoplasty, and the posterior membranous wall of the trachea and main bronchi are dissected to expose the airway. Marlex mesh is fashioned into a 2.0-cm-wide strip and sutured to the posterior membranous wall, reinforcing it. The graft is usually placed proximally from the level of the thoracic inlet and distally to the level of the mainstem bronchi. This procedure is offered to patients with severely symptomatic diffuse tracheobronchomalacia who are good surgical candidates and in whom a trial of central airway stenting has resulted in improved clinical symptoms. It is important to note that the use of long-term metallic stents in patients with diffuse tracheobronchomalacia should be avoided because of an increased risk of stent fracture due to changes in airway size and shape during respiration [14]. Silicone stents are preferable to metallic stents for treating malacia and also have the advantage of being easily removable [14]. Our results show that tracheoplasty resulted in a decrease in the degree of airway collapse in all five patients with postsurgical CT scans. Importantly, this quantitative improvement was accompanied by a qualitative improvement of respiratory symptoms. Changes in the tracheal shape resulting from the surgical procedure were also clearly verified by CT. Before surgery, the trachea was

biconvex (fishmouth) at end-inspiration in 60% of the patients, and it was lunate at dynamic expiration in 80% of the patients. This appearance can be explained by the pathophysiology of the disease: The flaccid posterior membranous wall of the trachea bows anteriorly during dynamic expiration in response to elevated intrathoracic extratracheal pressure [15]. Surgical reinforcement of the posterior membranous wall with tracheoplasty enhances the rigidity of this structure and makes it less susceptible to bowing during expiration. After tracheoplasty, 100% of the patients showed a normal-shaped trachea at end-inspiration and 80% showed a normal-shaped trachea at dynamic expiration. We realize that there are limitations regarding the use of dynamic CT for evaluating tracheobronchomalacia, particularly because this technique relies on the ability of the patient to cooperate with breathing instructions. For example, this factor was the cause for the one false-negative preoperative CT study in this series, which occurred in a patient with severe asthma. However, we emphasize that all other cases in our study showed a good correlation between dynamic expiratory CT and bronchoscopy; moreover, a good correlation between dynamic CT and bronchoscopy has also been described in the literature [3]. We also realize that there is a potential for measurement error in our methodology. We emphasize that the consensus review of mea-

1447

Downloaded from www.ajronline.org by 37.44.207.55 on 01/19/17 from IP address 37.44.207.55. Copyright ARRS. For personal use only; all rights reserved

Baroni et al.

A

B

Fig. 3.—62-year-old man (patient 1 in Table 1) with tracheobronchomalacia (same patient as shown in Fig. 1). A, Unenhanced CT image obtained 4 days after tracheoplasty shows large amount of fluid and soft-tissue density (arrow) surrounding trachea, corresponding to postoperative fluid and blood. B, Unenhanced CT image obtained 45 days after tracheoplasty shows postsurgical soft-tissue density has significantly decreased, with residual posterior airway wall thickening related to procedure (arrow).

surements by a second radiologist limited the potential for measurement error in our study. We did not attempt to validate the cross-sectional area measurements of the airways obtained with our tracing software by comparing them with phantom measurements because these methods have been previously validated by other authors using similar measurement software programs [16]. Despite these limitations, we have found that CT plays several potentially important roles in the pre- and postoperative settings. Its preoperative roles include the precise characterization of airway shape and determination of which parts of the airway wall contribute to excessive airway collapsibility. For example, patients with a lunate trachea at expiration are considered excellent candidates for tracheoplasty because this configuration implies that a redundant, “floppy” posterior membranous wall is the major factor contributing to the excessive airway collapse [4]. Thus, reinforcing the posterior membranous wall with a graft is likely to correct malacia in these patients. On the other hand, patients with normal posterior membranous walls and excessive collapsibility of other airway wall components (anterior, lateral) are unlikely to benefit from this procedure. The second role of preoperative CT is the evaluation for airway wall thickening and calcification. In combination with airway malacia, these findings suggest polychondritis, a disorder that is not treated surgically. The third

1448

role of preoperative CT is the evaluation for extrinsic paratracheal masses, which may preclude surgery. Finally, preoperative CT scans can serve as a baseline measure of airway collapsibility with which to compare postoperative scans for evaluating response to surgery. In the postoperative setting, CT provides a noninvasive method for assessing for postoperative complications and noninvasively quantifying the degree of improvement in airway collapsibility. With regard to the time interval for obtaining a CT scan postoperatively, as observed in one of our patients, CT scans in the early postoperative period (Fig. 3) can show transient postoperative changes in the mediastinum, such as fluid and blood, that should not be considered as complications related to the procedure. We thus recommend a minimum delay of 1 month after surgery before a final evaluation of the surgical results can be achieved. Obviously, if there is a concern for a postoperative complication, CT can be performed earlier. Finally, CT has the unique ability to visualize the characteristic thickening of the posterior wall of the airways after surgery, a finding observed in all our patients. This finding may serve as a clue for one to suspect that tracheoplasty has been performed. In our preliminary experience, our surgeons and pulmonologists have found a combination of subjective symptomatic improvement and quantitative reduction in airway collapsibility on CT to be the most helpful measurements of determining

response to surgery. Because not all patients in our study population underwent pre- and postoperative pulmonary function tests, we did not attempt to compare CT findings with pulmonary function test results. Future studies comparing CT and pulmonary function tests will be necessary to determine the precise contribution of these methods to determining response to therapy. In conclusion, dynamic expiratory CT is a potentially valuable tool in the pre- and postoperative evaluations of patients undergoing tracheoplasty, a novel surgical method that is increasingly being used for the definitive treatment of patients with severely symptomatic tracheobronchomalacia.

References 1. Johnson TH, Mikita JJ, Wilson RJ, Feist JH. Acquired tracheomalacia. Radiology 1973;109: 577–580 2. Heussel CP, Hafner B, Lill J, Schreiber W, Thelen M, Kauczor H-U. Paired inspiratory/expiratory spiral CT and continuous respiration cine CT in the diagnosis of tracheal instability. Eur Radiol 2001;11:982–989 3. Gilkeson RC, Ciancibello LM, Hejal RB, Montenegro HD, Lange P. Tracheobronchomalacia: dynamic airway evaluation with multidetector CT. AJR 2001;176:205–210 4. Wright CD. Tracheomalacia. Chest Surg Clin N Am 2003;13:349–357 5. Stern EJ, Graham CM, Webb WR, Gamsu G. Normal trachea during forced expiration: dynamic CT measurements. Radiology 1993;187: 27–31

AJR:184, May 2005

Downloaded from www.ajronline.org by 37.44.207.55 on 01/19/17 from IP address 37.44.207.55. Copyright ARRS. For personal use only; all rights reserved

Dynamic CT of the Central Airways 6. Gamsu G, Webb WR. Computed tomography of the trachea: normal and abnormal. AJR 1982;139:321–326 7. Webb EM, Elicker BM, Webb WR. Using CT to diagnose nonneoplastic tracheal abnormalities: appearance of the tracheal wall. AJR 2000;174: 1315–1321 8. Jokinen K, Palva T, Nuutinen J. Chronic bronchitis: a bronchologic evaluation. J Otorhinolaryngol 1976;38:178–186 9. Palombini BC, Villanova CA, Araujo E, et al. A pathogenic triad in chronic cough: asthma, postnasal drip syndrome and gastroesophageal reflux disease. Chest 1999;116:279–284 10. Hasegawa I, Boiselle PM, Raptopoulos V, Hatabu

H. Tracheomalacia incidentally detected on CT pulmonary angiography of patients with suspected pulmonary embolism. AJR 2003;181:1505–1509 11. Filler RM. Current approaches in tracheal surgery. Pediatr Pulmonol Suppl 1999;18:105–108 12. Herzog H. Relaxation and expiratory invagination of the membranous portion of the intrathoracic trachea and the main bronchi as cause of asphyxial attacks in bronchial asthma and the chronic asthmoid bronchitis of pulmonary emphysema [in German]. Schweiz Med Wochenschr 1954;84: 217–219 13. Rainer WG, Newby JP, Kelble DL. Long term results of tracheal support for emphysema. Dis Chest 1968;53:765–772

14. Boiselle PM, Feller-Kopman D, Ashiku S, Weeks D, Ernst A. Tracheobronchomalacia: evolving role of dynamic multislice helical CT. Radiol Clin North Am 2003;41:627–636 15. Kao SC, Smith WL, Sato Y, Franken EA, Kimura K, Soper RT. Ultrafast CT of laryngeal and tracheobronchial obstruction in symptomatic postoperative infants with esophageal atresia and tracheoesophageal fistula. AJR 1990;154:345–350 16. Hein E, Rogalla P, Hentschel C, Taupitz, M, Hamm B. Dynamic and quantitative assessment of tracheomalacia by electron beam tomography: correlation with clinical symptoms and bronchoscopy. J Comput Assist Tomogr 2000;24:247–252

Mark your calendar for the following ARRS annual meetings: May 15–20, 2005—New Orleans Hilton Riverside, New Orleans, LA; April 30–May 5, 2006—Vancouver Convention and Exhibition Centre, Vancouver, BC, Canada; May 6–11, 2007—Grande Lakes Resort, Orlando, FL; April 13–18, 2008—Marriott Wardman Park Hotel, Washington, DC.

AJR:184, May 2005

1449

Suggest Documents