This Article Abstract Full Text (PDF) Online Supplement All Versions of this Article: 200602-259OCv1 174/1/26    most recent Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Noppen, M. Articles by Vincken, W. Search for Related Content PubMed PubMed Citation Articles by Noppen, M. Articles by Vincken, W.

Fluorescein-enhanced Autofluorescence Thoracoscopy in Patients with Primary Spontaneous Pneumothorax and Normal Subjects

Interventional Endoscopy Clinic and Respiratory Division, and Anaesthesiology Department, University Hospital AZ-VUB, Brussels, Belgium; Critical Care and Respiratory Division, University of Athens, "Evaggelismos" Hospital, Athens, Greece; Division of Pulmonary and Critical Care Medicine, Queen's University, Kingston, Ontario, Canada; and Pulmonary Special Procedures Unit, Respiratory Division, University of California–San Diego, San Diego, California

Correspondence and requests for reprints should be addressed to Marc Noppen, M.D., Ph.D., Head, Interventional Endoscopy Clinic, University Hospital AZ VUB, 101 Laarbeeklaan, B 1090 Brussels, Belgium. E-mail: marc.noppen{at}az.vub.ac.be

	ABSTRACT

TOP ABSTRACT METHODS RESULTS DISCUSSION REFERENCES

 
Rationale: The exact site(s) and pathophysiology of air leakage in patients with primary spontaneous pneumothorax (PSP) are unknown. In one patient with PSP, fluorescein-enhanced autofluorescence thoracoscopy (FEAT) has shown areas of parenchymal abnormality unnoticed during white light thoracoscopy (WLT).

Objectives: To prospectively perform and compare WLT and FEAT in patients with spontaneous pneumothorax and in normal subjects.

Methods: One-time FEAT and WLT inspection with systematic mapping of semiquantified lesions in 12 consecutive patients with PSP was compared with one-time FEAT and WLT during sequential bilateral thoracoscopy in 17 control subjects.

Results: WLT abnormalities (anthracosis, cobblestone malformation, and blebs/bullae) were more prevalent in PSP. FEAT, however, showed high-grade lesions in PSP only, which often were present at areas that were normal, or that only showed anthracosis at WLT. When blebs/bullae were present, bleb-associated FEAT abnormalities were only present in two. Actual fluorescein leakage was seen in two patients with PSP.

Conclusions: Lungs in patients with PSP show significantly more abnormalities at WLT when compared with normal subjects. High-grade FEAT lesions were exclusively present in PSP, and predominantly at lung zones that appeared normal at white light inspection. These findings suggest that significant parenchymal abnormalities are not limited to lesions visible during WLT, such as blebs and bullae.

Key Words: fluorescein • spontaneous pneumothorax • thoracoscopy

Although there is agreement on the timing of, and indication for, recurrence prevention treatment in primary spontaneous pneumothorax (PSP) (1, 2), there is no consensus on the optimal technique for achieving recurrence prevention. Pulmonologists only perform pleurodesis, using medical thoracoscopy with talc poudrage or with pleural abrasion, whereas surgeons usually advocate more invasive procedures such as video-assisted thoracoscopic bleb- or bullectomy followed by a pleurodesis procedure such as talc poudrage, pleural abrasion, or partial pleurectomy. Both approaches show comparable efficacy in historical, noncomparative studies (3–5).

This difference in treatment approach is a reflection of the differences in opinion on the role of visible lung abnormalities (i.e., blebs and/or bullae) in the pathogenesis of PSP (6–8). Although most authors believe that rupture of blebs or bullae inevitably is the cause of PSP, which supports the surgical concept of systematic bleb/bulla treatment by stapling or suturing, we (3, 4, 6, 7) and others (9, 10) have proposed that air leakage may occur at other diseased areas of the lung. These areas may be characterized by visceral pleural porosity, a process that can be present at blebs or bullae but also at macroscopically normal parts of the lung (11, 12).

To try to visualize abnormal areas at the visceral lung surface and identify sites of potential air leakage, we conceptualized fluorescein-enhanced autofluorescence thoracoscopy (FEAT) (13). Sodium fluorescein is a relatively low molecular weight, highly water-soluble compound, which, when exposed to light of wavelengths between 465 and 490 nm (blue light, which is the wavelength used by the DAFE autofluorescence system), emits light at a wavelength of 520 to 530 nm (green-yellow, which also is detectable by the DAFE system). Fluorescein is traditionally used in the diagnosis and treatment of ocular disorders. Its visible fluorescence on leakage from damaged vessels makes it useful in the diagnosis of retinal vascular disorders and monitoring of treatment of conditions amenable to laser photocoagulation (14). In more than 30 yr of clinical use, the intravenous injection of fluorescein has been proven safe, although mild side effects occasionally have been reported. Serious or life-threatening complications are extremely rare.

We previously reported on a patient with PSP (13) in whom a fluorescein aerosol was administered, followed by sequential white light and blue light thoracoscopy using a blue light–operating DAFE autofluorescence system originally designed and used for autofluorescence bronchoscopy. Several areas of important subpleural fluorescein accumulation, undetectable by normal white light thoracoscopy (WLT), were observed. This observation supported the concept that air leakage in PSP is not necessarily associated with, or located at, blebs or bullae but can occur at areas that have a completely normal appearance during WLT.

We therefore hypothesize that FEAT, when systematically performed in patients with PSP, may show parenchymal abnormalities that are undetected by WLT, which are not necessarily located at blebs and bullae, and which are absent in normal subjects. To confirm or refute this hypothesis, we conducted a prospective, controlled study using one-time sequential WLT and FEAT in a cohort of consecutive patients with recurrent or persistent PSP, as well as in a cohort of control subjects consisting of otherwise healthy, age-matched subjects undergoing bilateral thoracoscopic sympathicolysis for autonomic disorders.

	METHODS

TOP ABSTRACT METHODS RESULTS DISCUSSION REFERENCES

 
Patients
During a 6-mo period, all consecutive patients presenting with recurrent and/or persistent PSP, hence candidates for thoracoscopic treatment, were included in the study. The control group, recruited during the same period, consisted of age-matched subjects suffering from autonomic disorders (predominantly essential hyperhidrosis) for which bilateral thoracoscopic sympathicolysis was proposed. Because the effect of inhaled fluorescein on fetal development is unknown and because of the male predominance in PSP, only male control subjects were recruited. Ethics committee approval (ref. 2004/086D) and oral informed consent was obtained from all subjects before enrollment.

Methods
Approximately 10 to 30 min before anesthesia induction, all subjects inhaled an aerosolized 10% fluorescein solution (5 ml; Novartis Pharma, Vilvoorde, Belgium) for approximately 10 min under normal tidal volume conditions. The fluorescein aerosol was delivered via a pressure-driven nebulizer (Micro Mist; Hudson RCI, Vasby, Sweden) attached to a mask. All patients underwent total intravenous anesthesia, with single-lumen intubation with high-frequency jet ventilation delivered through the endotracheal tube. Thoracoscopy was performed using a 7-mm trocar for introduction of the telescope, and a 5-mm trocar for introduction of accessories. We used the DAFE autofluorescence endoscopy unit (Richard Wolf, Knitlingen, Germany) equipped with a filtered 300-W xenon lamp with an IR blocking filter allowing for inspection with either white light or violet-blue light (ranging from 390 to 460 nm) as the excitation source. The light is transmitted from the lamp to the thoracoscope via a liquid light guide. Photodetection relies on the use of a CCD camera and a dual detection range: 500 to 590 nm and 600 to 700 nm. A system of dichroic mirrors and filters allows for superimposition of the signals and false color image reconstruction is displayed on a color monitor (15). Switching from white to blue light and back is accomplished instantaneously by means of a foot pedal switch.

Thoracoscopy was always started in white light conditions; after a systematic and complete inspection and recording of abnormalities, a similar inspection was performed in blue light conditions. White light and blue light observations were systematically categorized semiquantitatively (Table 1, and online supplement). In the control subjects, both lungs were inspected and the data were pooled. After inspection, patients with PSP were treated with thoracoscopic talc poudrage, regardless of eventual abnormal white or blue light findings (except for Case 2 who was referred for surgical correction of severe vascular abnormalities); control subjects underwent sympathicolysis, as described elsewhere (16).

	RESULTS

TOP ABSTRACT METHODS RESULTS DISCUSSION REFERENCES

 
Twelve consecutive patients with recurrent and/or persistent (> 4 d air leakage) PSP were included in a 6-mo period. PSP was defined as a spontaneously occurring pneumothorax in a patient without known underlying lung disease. Mean age was 32.6 ± 9.6 yr; range, 20–49 yr. There were 10 males, and 10 of 12 patients (83.3%) were smokers. During the same study period, 17 control subjects were recruited. All control subjects were referred for bilateral thoracoscopic sympathicolysis for treatment of autonomic disorders (essential hyperhidrosis, 16 cases; idiopathic pathologic flushing/social phobia syndrome, 1 case). Mean age was 29.5 ± 9 yr; range, 20–49 yr (p = 0.3). Of the 17 control subjects, 5 (29.4%) were smokers (p = 0.008).

All thoracoscopies were uneventful. The addition of blue light inspection prolonged total procedure time by only a few minutes. In the control group, one FEAT inspection could not be performed because of technical failure of the camera; hence, only 33 lung inspections were recorded. All control subjects left the hospital on the day of the procedure, and all but one patient with PSP left the hospital within 72 h. One patient with PSP (Case 2) was referred to thoracic surgery, and eventually underwent a left thoracotomy for treatment of dramatic vascular malformations at the left upper lobe (hospitalization, 8 d); all other patients underwent talc poudrage. One patient (Case 12) had persistent air leak after talc poudrage, and was dismissed with a 16-F chest tube attached to a Heimlich valve. The drain was removed 8 d after thoracoscopy. No patient has presented with recurrence in a 6 to 12 mo follow-up period. Results of WLT and FEAT observations are shown in Tables 2 and 3.

To our surprise, FEAT inspection unexpectedly yielded a greater prevalence of low-grade (+ and ++) lesions in control subjects (27/33 or 82%) than in PSP lungs (5/12 or 42%; difference, 40% 95% CI for difference, 9–71%; Fisher's exact test, p = 0.022). However, high-grade lesions (+++ and ++++) were more often observed in PSP lungs (7/12 or 58%) as compared with control lungs (0%; 58% difference; 95% CI for difference, 31–86%; Fisher's exact test, p < 0.0001). In the majority of cases, high-grade FEAT lesions were present at areas that were normal or only showed anthracosis lesions at WLT. When blebs/bullae were present in PSP (six cases), there were no bleb-associated FEAT abnormalities present in four (67%); in five of six cases with PSP with blebs /bullae, FEAT abnormalities were only present elsewhere in the lung. Actual fluorescein leakage was observed in two patients with PSP and with blebs/bullae; in one case this was not at the site of the blebs; in one case this was near, but not at, a bulla.

	DISCUSSION

TOP ABSTRACT METHODS RESULTS DISCUSSION REFERENCES

 
This is the first prospective, controlled study of in vivo visualization of abnormal lung regions that may play an important role in the pathophysiology of PSP by the use of FEAT, comparing white light and FEAT findings in PSP and normal control subjects. PSP lungs showed significantly more abnormalities at WLT when compared with normal lungs, including the presence of blebs/bullae. FEAT showed low-grade lesions in the majority of normal lungs, but high-grade lesions (extensive subpleural fluorescein accumulation and fluorescein leakage) were exclusively present in PSP. Interestingly, these high-grade lesions (which may be considered as possible sites where air leakage can originate) were not necessarily associated with blebs or bullae or other abnormalities visible at white light inspection. These findings confirm the observations in a previous first case report (13), which led to the current study hypothesis, and suggest that the observed high-grade FEAT phenomena (which are limited to PSP lungs only) may be proof of the presence of genuine lung parenchymal abnormalities elsewhere than at blebs or bullae, the typical culprits of air leakage location in PSP. Nevertheless, the pathophysiologic significance of the observed phenomena remains to be explored, while acknowledging the limitations inherent to this technique.

First, FEAT data interpretation can only be realized in a semiquantitative fashion, because fluorescence measurements are planar measurements with light intensity that not only depend on the local concentration of the fluorophore but also on how deep the fluorophore penetrates the lung tissue (17). In addition, the shape of the emitted light spectrum may be affected by a differential absorption of different frequencies (15). FEAT lesions are classified as absent (0), + (discrete, greenish, poorly defined subpleural glow), ++ (distinctive, well-localized spots of clear greenish fluorescence), +++ (areas of intense fluorescence, well localized), and ++++ (visible fluorophore leakage from the lung).

Second, the semiquantitative scoring is entirely unblinded, because the thoracoscopist is in full knowledge of whether the patient has had a pneumothorax or is a control subject. Unconscious bias in reporting may furthermore be increased by the fact that a subjective, semiquantitative scoring system was used. The risk for bias was minimized (although not abolished) by simultaneous scoring by the principal thoracoscopist (M.N.) and by one of the fellows (T.D., S.H., G.S., K.A., or P.M.) assisting the procedure, and by the fact that every procedure was videotaped and reviewed after the procedure. In the future, improved scorer blinding and/or a more objective method of fluorescence measurement will be necessary.

Third, the pathophysiologic significance and pathologic correlate of the observed abnormalities are unknown. The kinetics of inhaled fluorescein with respect to absorption in the bloodstream, interstitium, or cells are unknown. We hypothesize that the high-grade +++ lesions (which are exclusively seen in PSP and not in control lungs) may be due to inhaled fluorescein aerosol accumulation in certain areas of the lung due to localized areas of air trapping, which is one of the plausible mechanisms leading to spontaneous air leakage and pneumothorax (3). Interestingly, this interpretation can be brought in agreement with the observation of a wedge-shaped region of increased helium signal in the lung parenchyma subjacent to the site of air leakage in an animal model of pneumothorax, using three-dimensional magnetic resonance imaging of laser-polarized ³He gas (18). Alternatively, fluorescein may accumulate in areas that receive more rather than less ventilation; only the observation of the wash-in and wash-out phases of fluorescein administration in a patient or animal model would help clarify the true meaning of fluorescein accumulation. In any case, the fluorescein clearly accumulates under the visceral pleural surface of the lung, because it cannot be washed away or aspirated during thoracoscopy. The ++++ lesions (actual fluorophore leakage through the visceral pleura) also are exclusively seen in PSP. In the absence of blebs, bullae, or other abnormalities at white light inspection, this phenomenon can be explained by the presence of pleural porosity (the presence of pleural pores several microns in width) (19). The exact nature of this pleural porosity also is unknown and a possible subject of further study. Loss of surface mesothelial cells, thinning and rupture of the basement membrane, and/or down-regulation of junctional proteins may play a role. It is possible that an interplay or peripheral airway and airspace inflammation (which may be enhanced by smoking), mechanical factors (more negative pressures around, and increased mechanical stretching of the lung apices in ectomorphic individuals), and changes in collagen quality may lead to structural changes qualified as "pleural porosity." Either the aerosol itself (with 78.5% of its mass in the < 5 µm range for the Micro Mist nebulizer [Hudson RCI]) could have passed right through, or the fluid molecules (once the fluorescein aerosol droplets deposit) may have penetrated preferentially in the more porous pleural areas, with a detectable fluorescence intensity as a result. The presence of fluorescein-containing foam on the visceral pleural surface, in the absence of blebs or bullae, may be due to the high solubility of fluorescein in water, which favors its spread over the parenchymal surface around the leak. Nevertheless, the anatomopathologic correlate of our observations remains to be determined and is a topic of ongoing research.

To our knowledge, inhalation of fluorescein has not yet been reported in controlled series, although it has been used safely in the past under WLT conditions (20). The results of fluorescein inhalation in white light illumination conditions for detecting air leaks, however, are disappointing (C. Boutin, M. Noppen, unpublished observations). Illumination with blue light, taking advantage of the physical fluorescence properties of fluorescein, seems to dramatically increase its sensitivity, as this study illustrates.

Despite the limitations and unanswered questions from this study, our findings are intriguing. Indeed, abnormal regions that may be susceptible to leakage, or that actually show visceral pleural air leakage, can be identified with FEAT in the absence of blebs and bullae, and in regions that appear completely normal at WLT. This suggests that patients with PSP may suffer from a more extensive disease process of parenchymal inflammation and destruction than previously believed, and challenges the idea that bleb or bulla rupture is involved in the pathogenesis of every case of PSP. In this respect, and when confirmed and corroborated by pathologic examinations, FEAT may have therapeutic implications in the future because it may allow for in vivo identification of all diseased lung regions and, if they are still present at the time of thoracoscopy, of actual air leaks. Our observations indeed may explain why bleb resection alone (blebs identified by WLT) is followed by a 5 to 10% recurrence rate, whereas a pleurodesis technique alone, or in association with bleb resection, decreases the recurrence rate to 0 to 5% (3). Our observations of widespread abnormalities in all patients with PSP strengthen the concept that not the systematic resection of diseased areas (as identified by conventional WLT) but pleurodesis is the cornerstone of successful recurrence prevention (3).

Finally, FEAT is well tolerated and safe (no side effects except for an asymptomatic yellow discoloration of the mouth, which resolves within 24–48 h). It is easy to perform, with pedal-switch alternation between white and blue light inspection, prolonging thoracoscopy by a negligible amount. Although the DAFE autofluorescence endoscopy unit has been developed for the bronchoscopic detection of early or preneoplastic bronchial lesions (15), and is easily adaptable to commercially available bronchoscopes, it can also easily be adapted to the rigid telescope used in thoracoscopy.

In conclusion, WLT shows significantly more lung abnormalities in patients with PSP as compared with normal control subjects. Addition of FEAT yields significantly more regions of abnormal fluorescence in patients with PSP, and furthermore allows identifying regions that appeared normal during WLT. These findings may have a significant impact on the study of the pathophysiology of PSP, and on therapeutic algorithms in PSP recurrence prevention.

	FOOTNOTES

 
This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org

Originally Published in Press as DOI: 10.1164/rccm.200602-259OC on April 20, 2006

Conflict of Interest Statement: None of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript.

Received in original form February 21, 2006; accepted in final form April 20, 2006

	REFERENCES

TOP ABSTRACT METHODS RESULTS DISCUSSION REFERENCES

 

1. Miller AC. Spontaneous pneumothorax. In: Light RW, Lee YCG, editors. Textbook of pleural diseases. London: Arnold; 2003. pp. 445–463.
2. Light RW. Management of spontaneous pneumothorax. Am Rev Respir Dis 1993;148:245–248.[Medline]
3. Noppen M, Baumann MH. Pathogenesis and treatment of primary spontaneous pneumothorax: an overview. Respiration (Herrlisheim) 2003;70:431–438.
4. Noppen M, Schramel F. Pneumothorax. Eur Respir Rev 2002;7:279–296.
5. Baumann MH, Noppen M. Pneumothorax. Respirology 2004;9:157–164.[CrossRef][Medline]
6. Noppen M. Do blebs cause spontaneous pneumothorax? Con: blebs do not cause spontaneous pneumothorax. J Bronchol 2002;9:319–325.
7. Noppen M. Management of primary spontaneous pneumothorax: does cause matter? Monaldi Arch Chest Dis 2001;56:344–348.[Medline]
8. Baumann MH. Do blebs cause spontaneous pneumothorax? Pro: blebs cause spontaneous pneumothorax. J Bronchol 2002;9:313–318.
9. Tschopp JM, Bolliger CT, Boutin C. Treatment of spontaneous pneumothorax: why not simple talc pleurodesis by medical thoracoscopy? Respiration (Herrlisheim) 2000;67:108–111.[CrossRef][Medline]
10. Smit HJ, Wienk MA, Schreurs AJ, Schramel FM, Postmus PE. Do bullae indicate a predisposition to recurrent pneumothorax? Br J Radiol 2000;73:356–359.[Abstract]
11. Radomsky J, Becker HP, Hartel W. Pleuraporosität beim idiopatischen Spontanpneumothorax [in German]. Pneumologie 1989;43:250–253.[Medline]
12. Ohata M, Suzuki M. Pathogenesis of spontaneous pneumothorax: with special reference to the ultrastructure of emphysematous bullae. Chest 1980;77:771–776.[Abstract/Free Full Text]
13. Noppen M, Stratakos G, Verbanck S, D'Haese J, Meysman M, Vincken W. Fluoresceine enhanced autofluorescence thoracoscopy in primary spontaneous pneumothorax. Am J Respir Crit Care Med 2004;170:680–682.[Abstract/Free Full Text]
14. Rechtman E, Harris A, Kumar R, Cantor LB, Ventraprogada S. Desai M, Friedman S, Kagemann L, Garzozi HJ. An update on retinal circulation assessment techniques. Curr Eye Res 2003;27:329–343.[Medline]
15. Goujon D, Zellweger M, Radu A, Grosjean P, Weber BC, van den Bergh H, Monnier Ph, Wagnières G. In vivo autofluorescence imaging of early cancers in the human tracheobronchial tree with a spectrally optimized system. J Biomed Opt 2003;8:17–25.[Medline]
16. Noppen M, Herregodts P, D'Haese J, D'Haens J, Vincken W. A simplified T2–T3 thoracoscopic sympathicolysis technique for the treatment of essential hyperhidrosis: short-term results in 100 patients. J Laparoendosc Surg 1996;6:151–159.[Medline]
17. Schuster DP, Kovacs A, Garbow J, Piwnica-Worms D. Recent advances in imaging the lungs in intact small animals. Am J Respir Cell Mol Biol 2004;30:129–138.[Abstract/Free Full Text]
18. Roberts DA, Rizi RR, Lipson DA, Aranda M, Baumgardner J, Bearn L, Hansen-Flaschen J, Gefter W, Hatabu HH, Leigh JS, et al. Detection and localization of pulmonary air leaks using laser-polarized 3He MRI. Magn Reson Med 2000;44:379–382.[CrossRef][Medline]
19. Becker HP, Schmitz I, Radomsky J, Muller KM. Ultrastrukturelle Befunde: pleuraler Cysten beim Spontanpneumothorax [in German]. Chirurg 1995;66:787–793.[Medline]
20. Boutin C, Viallat JR, Aelony Y. Practical thoracoscopy. Berlin and Heidelberg, Germany: Springer-Verlag; 1991.

This article has been cited by other articles:

				J-M. Tschopp, J-M. Schnyder, M. Froudarakis, and P. Astoul VATS or simple talc poudrage under medical thoracoscopy for recurrent spontaneous pneumothorax Eur. Respir. J., February 1, 2009; 33(2): 442 - 443. [Full Text] [PDF]

				M. H. Baumann To Bleb or Not To Bleb? Chest, October 1, 2007; 132(4): 1110 - 1112. [Full Text] [PDF]

				T.-W. Huang, S.-C. Lee, Y.-L. Cheng, C. Tzao, H.-H. Hsu, H. Chang, and J.-C. Chen Contralateral Recurrence of Primary Spontaneous Pneumothorax Chest, October 1, 2007; 132(4): 1146 - 1150. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Online Supplement All Versions of this Article: 200602-259OCv1 174/1/26    most recent Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Noppen, M. Articles by Vincken, W. Search for Related Content PubMed PubMed Citation Articles by Noppen, M. Articles by Vincken, W.