LABORATORY INVESTIGATION

A Pilot Study on the Feasibility of InterventionalLung Volume Reduction

Wansheng Wang, Yonghua Dong, Bin Liu, Chi Zhu, Yongqiang Yu

Department of Radiology, First Affiliated Hospital of Anhui Medical University, Hefei, Anhui, China, 230022

Abstract

The objective of this study was to evaluate the feasibility andsafety of lung volume reduction by transbronchial alcoholand lipiodol suspension infusion with the aid of balloon-tipped catheter occlusion. Twenty-six healthy adult rabbitswere divided into four treatment groups: alcohol and lipiodolsuspension infusion (n = 8), lipiodol infusion (n = 8), alco-hol infusion (n = 5), or bronchial lumen occlusion (n = 5).After selective lobar or segmental bronchial catheterizationusing a balloon-tipped occlusion catheter, the correspondingdrug infusion was performed. Bone cement was used toocclude the bronchial lumen in the occlusion group. Theanimals were followed up for 10 weeks by chest X-ray andcomputed tomography (CT), and then the whole lungs wereharvested for histological examination. Alcohol and lipiodolsuspension or lipiodol could be stably retained in alveoli inthe first two groups based on chest X-ray and CT, butobvious collapse only occurred in the group receiving alco-hol and lipiodol suspension or the bronchial lumen occlusiongroup. Histological examination revealed damage and dis-ruption of the alveolar epithelium and fibrosis in related lungtissue in the group receiving alcohol and lipiodol suspension.Similar changes were seen in the bronchial lumen occlusiongroup, apart from obvious marginal emphysema of the targetareas in two animals. Interstitial pneumonia and dilatedalveoli existed in some tissue in target areas in the lipiodolgroup, in which pulmonary fibrosis obliterating alveoli alsooccurred. Chronic alveolitis and pleural adhesion in targetareas occurred in the group infused with alcohol alone,whereas visceral pleura of the other three groups was regularand no pleural effusion or adhesion was found. Alcohol andlipiodol suspension that is stably retained in alveoli can re-sult in significant lung volume reduction. Through alcoholand lipiodol suspension infusion, obstructive emphysema orpneumonia arising from bronchial lumen occlusion could beavoided.

Key words: Lung volume reduction—Alcohol—Lipiod-ol—Pulmonary fibrosis

Lung volume reduction therapy refers to the elimination ofemphysematous hyperinflated lung through surgical means(lung volume reduction surgery, LVRS) or lung volumereduction through minimally invasive techniques, whichallows the remaining pulmonary parenchyma and therespiratory muscles to function more effectively [1–13].Although the physiologic and symptomatic benefits ofLVRS have been quite impressive, substantial morbidityand mortality, linked with major thoracic surgery in anelderly debilitated population, have limited the clinicalutility of LVRS [1–7]. However, the demonstrated benefitsof LVRS and the rigors of surgical lung volume reductionhave led investigators to develop minimally invasive tech-niques to achieve lung volume reduction. One-way valveimplant, which is aimed at reducing lung volume by causingatelectasis of the target area, had a lower achievement ratiobecause target areas communicated with surrounding lungtissue through collateral passages [8–11]. Ninety-one per-cent of the target areas showed lung contract and scar tissuein Ingenito�s experiment, but the safety issues of fibrin gluerequire close attention [12, 13]. Lung volume reductions byalcohol and lipiodol suspension infusion or Pingyangmycinand lipiodol emulsion infusion followed by occlusion of thebronchial lumen with bone cement were tested effectively tosome extent in studies in China [14–16]. Pingyangmycincan cause pulmonary fibrosis in target areas while the samechange might occur in nontarget areas, so it is possible toimpair gas-exchange function in patients with emphysema[17, 18]. Our previous study [19] suggested that bonecement could not be used in the clinical setting because it ishard to control. It is difficult to make the drug distribute inthe target uniformly, thus, obstructive emphysema, atelec-tasis, or inflammation of adjacent lung tissue in target areasmight occur once related bronchial lumens were occludedwith bone cement. Although lipiodol could stably remain inalveoli and result in positive fibrosis, it could not cause

Correspondence to: Y. Q. Yu and Y. H. Dong; email: shenyuxi@mail.hf.ah.cn; email: dongyhua@yahoo.com

ª Springer Science+Business Media, LLC 2007Published Online: 23 June 2007CardioVascular

and InterventionalRadiology

Cardiovasc Intervent Radiol (2008) 31:177–182

DOI: 10.1007/s00270-006-0211-8

obvious reductions in lung volume without occlusion of theproximal bronchial lumen. Therefore, an attempt was madeto find a new mini-invasive technique by which conspicuousreduction of lung volume could be achieved withoutrequiring occlusion of bronchial lumens.

MethodsExperiments were conducted in accordance with a protocol ap-proved by the Animal Care and Use Committee, Anhui MedicalUniversity. All interventions were performed under light anes-thesia. Anesthesia was induced with intravenous administration of3% pentobarbital (1 mL/kg). A balloon-tipped occlusion catheterwas induced into the trachea under the guidance of digital sub-traction angiography (DSA) and was advanced to the lobar orsegmental bronchus of the target site. Heart rate and arterialoxygen saturation were continuously monitored using an oximeterand tongue probe.

Interventional Lung Volume Reduction Procedure

Twenty-six healthy adult rabbits were divided into four treatmentgroups: (1). alcohol and lipiodol suspension infusion (n = 8), (2)lipiodol infusion alone (n = 8), (3) alcohol infusion alone (n = 5),or (4) bronchial lumen occlusion (n = 5). After selective lobar orsegmental bronchi catheterization of a self-made balloon-tippedocclusion catheter (8 mm), the corresponding drug infusion wasperformed and the extra drug in the bronchial lumen was re-moved as much as possible to avoid the reflux of drugs. In thebronchial lumen occlusion group, bone cement was used to oc-clude bronchial lumens following alcohol and lipiodol suspensioninfusion (n = 2) or lipiodol infusion (n = 3). Posteoanterior (PA)and lateral chest X-ray were performed to follow changes in thetargeted area. The alcohol and lipiodol suspension was madefrom lipiodol and 75% alcohol by specific concentration (1 mL:1mL). The dosage of alcohol and lipiodol suspension or lipiodolgiven (2–3 mL) was standardized based on the observation ofalveoli under fluoroscopy. The dosage used in the group infusedwith alcohol alone was 2 mL. Bronchial lumens were occludedwith 1 mL of bone cement . The animals were maintained underroutine conditions after the procedure, and the eating and drink-ing status, activity, and vital signs in each animal were moni-tored.

Radiologic Assessments and Pathology

All animals were followed up for 10 weeks by PA and lateral chestX-ray and computed tomography (CT), which were carried out atday 1, day 3, week 1, week 4, and week 10 after completion oftreatment (Post-Rx). After 10 weeks, all animals were euthanizedwith intravenous pentobarbital (100 mg/kg) and autopsied. Thelungs were harvested for visual inspection along with the completetrachea, and photographs were taken (including PA and lateralchest X-ray films for the specimen). The lungs were sliced seriallyinto 2–2.5-mm axial sections and further inspected for signs of scarformation or other gross changes (infection, pleural adhesions).Samples from target sites and representative samples from non-target sites were fixed in 15% buffered formalin and stained withhemotoxylin and eosin. The degrees of collapse of target areas seen

by PA and lateral chest films were categorized into four grades:grade 0 (0% collapse), grade 1 (<25% collapse), grade 2 (25–75%collapse), and grade 3 (>75% collapse) according to the method ofYim [10].

Statistical Analysis

Statistical significance was defined as a p-value less than 0.05.Differences between specific groups were identified using therank-sum test. All statistical analysis was performed using SPSSfor Windows version 10.0.

Results

Drug Retention and General Condition

Alcohol and lipiodol suspension or lipiodol was primarilyretained in the central target areas (Figs. 1A–3A). Nobackflow to nontarget sites was seen immediately afterextubation. One animal in the occlusion group died frombone cement occlusion of the trachea lumen immediatelyafter extubation. The distribution of alcohol in the targetarea could not be observed because alcohol cannot bedemonstrated on chest radiography. Two rabbits in thegroup receiving alcohol infusion died 3 days and 6 weeksafter the procedure respectively, and the other two animalsin the same group developed extensive subcutaneousemphysema. The drinking and eating status, activity, andvital signs of other animals remained stable after the pro-cedure.

Radiologic Changes After the Procedure inDifferent Groups

In the alcohol and lipiodol suspension infusion group orocclusion group, target areas (most target sites located inposterior segment of lower lobe) collapsed gradually overtime and finally formed complete atelectasis with obviouscontraction (Figs. 1B and 2B). Thereafter, the drug in thealveoli was absorbed and the density of the target area de-creased gradually. In the lipiodol infusion group (Fig. 3B)and alcohol infusion group, no obvious contraction in thetarget areas was seen.

A small quantity of alcohol and lipiodol suspension orlipiodol deposited in nontarget sites with two animals in thealcohol and lipiodol suspension group and three animals inthe lipiodol group 1 day after the procedure. No newdeposition in nontarget areas was found in radiologic films 3days, 1 week, 4 weeks, and 10 weeks after treatment. Theextent of lung volume reduction in each group 10 weeksafter the procedure is shown in Table 1.

Pathologic Findings

Obvious contraction of target areas occurred in the alcoholand lipiodol suspension group or occlusion group. The tis-sue appearances in the groups were brown or brunneus,

178 W. Wang et al.: Feasibility of ILVR

smooth, and bright, and the tissue was flexible (Fig. 1C and2C). The margin of the target areas became obviousemphysema on two animals in the occlusion group(Fig. 4A).

In the groups receiving pure lipiodol or pure alcohol,target areas did not contract obviously, and they appeared asgray or brunneus along with pleural pachynsis and variablenumbers of adhesions.

Fig. 3. Changes in rabbit 9 duringand 10 weeks after ILVR. (A) Lipiodoldeposits in the target area during theoperation. (B) The target area doesnot exhibit obvious contraction 10weeks after treatment.

Fig. 2. Changes in rabbit 2 during and 10 weeks afterILVR. (A) Alcohol and lipiodol suspension deposits in de-cocted lateral segment of left lower lobe during the oper-ation. (B) The targert area (arrow) contracts obviously and

no high density is observed in other regions of lung 10weeks later. (C) The X-ray film of the specimen shows thatthe smaller triangular high density precisely outlines thetarget region.

Fig. 1. Changes in rabbit 1 during and 10 weeks afterinterventional lung volume reduction (ILVR). (A) The alcoholand lipiodol suspension deposits in target area during theoperation. (B) The target area contracts obviously and no

high density is observed in other regions 10 weeks later.(C) The X-ray film of the specimen 10 weeks later showsthat the triangular high density precisely outlines the targetregion.

W. Wang et al.: Feasibility of ILVR 179

Histological examination revealed damage and disrup-tion of the alveolar epithelium, severe atrophy, and fibrosisof the lung tissue in the group receiving the alcohol andlipiodol suspension (Fig. 5). Similar changes had been seenin the lumen occlusion group, apart from emphysema ofmarginal target areas in two animals. Interstitial pneumoniaand dilated alveoli were found in some tissue of target areasin the pure lipiodol group (Fig. 4B) compared with normalrabbit (Fig. 4C), and pulmonary fibrosis obliterating alveolioccurred in other areas. Chronic alveolitis of target areasoccurred in the group treated with pure alcohol. Eosinophilegranulocytes and lymphocytes exuded in interstitial regionsalong with numerous aggregating pulmonary foam cells(Fig. 6).

Discussion

The results of the present study demonstrated that alcoholand lipiodol suspension or lipiodol alone can stably remainin the alveoli of target sites without requiring occlusion ofbronchial lumens. The histology of alveoli is different fromthat of bronchus, as it lacks cough reflex arc and ciliatedepithelium. This is the reason why foreign bodies in thebronchus can be expectorated at the end and those in alveolicould not. However, the gravitational effect might be anegative factor that can help lead to drug drainage, becausemost target sites located in the posterior segment of thelower lobe, which occupies the superior position and whose

segmental bronchus debouches downward (shown clearly inlateral films). The results of our study have demonstratedthat alcohol and lipiodol suspension or lipiodol alone inalveoli typically could not be discharged resulting from thecough reflex or gravitational effect. This offers an experi-mental foundation for the further clinical application ofalcohol and lipiodol suspension infusion in any pulmonarysegments or lobes without concerns of occluding the bron-chial lumen.

Table 1. Summary of lung volume reduction on chest film after interventional lung volume reduction

GradeAlcohol and ipiodolsuspension infusion group

Lipiodol infusiongroup*

Alcohol infusiongroup*

Occlusiongroup

Grade 0 (0% collapse) 0 0 3 0Grade 1 (<25% collapse) 0 4 0 0Grade 2 (25–75% collapse) 2 4 0 3Grade 3 (>75% collapse) 6 0 0 1

Note: Compared to alcohol and lipiodol suspension infusion.*v2 = 16.518, p = 0.01.

Fig. 4. (A) Photograph of rabbit 21 specimen 10 weeksafter treatment shows obvious contraction of the target area,with brunneus and smooth appearance (bent arrow), andobvious marginal emphysema (arrow). (B) Hemotoxylin and

eosin (H&E) stain microscopy of marginal lung tissue withemphysema from the same rabbit shows obvious alveolarexpansion along with pneumatocele (·100). (C) H&E stain oflung tissue in a normal rabbit (·100).

Fig. 5. H&E stain of the target area of rabbit 1 showsdamage and disruption of the alveolar epithelium, severecollapse, and focal fibrosis in the group receiving alcohol andlipiodol suspension (·100).

180 W. Wang et al.: Feasibility of ILVR

Lung volume reduction by absolute alcohol and lipiodolsuspension injecton was carried out first in 2004 [14]. Intheir experiment, pneumonia of nontarget areas occurred intwo rabbits 1 week or 2 weeks after treatment, because ofnot occluding the bronchial lumen with bone cement. Theauthor suggested that absolute alcohol and lipiodol sus-pension injection should be combined with occlusion ofthe bronchial lumen for lung volume reduction. In thepresent study, the balloon-tipped occlusion catheter wasutilized, and the extra drug in the bronchial lumen wasremoved as much as possible, which avoided the reflux ofdrugs.

Pulmonary fibrosis resulted from the constant impair-ment of alveolar epithelial cell caused by drugs. Meanwhile,obvious lung volume reduction would appear once the scarpoint of attachment or pulmonary atelectasis exists in targetsites [12–17]. It was shown that superselective bronchiallipiodol infusion can lead to positive fibrosis, but lungvolume reduction is not satisfactory in the present study.Interstitial inflammation or fibrosis formed rather thanfibrosis obliterating alveoli, because scar attachment pointsdisappeared with the absorption of lipiodol. Thus, the effectof lung volume reduction was negatively influenced by theinfusion of lipiodol alone. However, alcohol and lipiodolsuspension infusion can result in obvious lung volumereduction. Widespread fibrosis formed on the background ofcomplete atelectasis in the target area. Pulmonary atelectasismight have been caused by potential inactivation of alveolarsurfactant by alcohol. The pulmonary atelectasis with col-lapse of alveoli offers a favorable circumstance for fibrosisobliterating alveoli. Lipiodol infusion with occlusion to thebronchial lumen also leads to obvious lung volume reduc-tion as a result of fibrosis obliterating alveoli. In the groupinfused with alcohol alone, obvious contract of target areasdid occur, and the complications associated with its use aretoo severe, so we conclude that alcohol infusion has littleclinical value.

Ideal lung volume reduction was obtained throughalcohol and lipiodol suspension infusion, the dosage ofwhich was standardized on the basis of the presence oflight shadow in target area under fluoroscopy in thepresent study. Lipiodol plays a marked role in defining thedegree to which the drug flows into alveoli. The dosage ofdrug is especially vital for lung volume reduction; fibrosiscould not form if less or no drug deposits in alveoli,whereas large-area oppressive necrosis in the target areacould occur if more drug was infused [12]. Obviously, itwas very difficult to control the dosage of drug in eachtarget area when drug infusion was performed underbronchoscope [12, 13]. It was shown in the present studythat high-density infusion of alcohol and lipiodol suspen-sion or lipiodol alone can clearly outline the target region.This approach makes it easy for us to control the dosageused in each target and is useful for following the changeof lung volume reduction.

Fibrin glue or bone cement was used to occlude thebronchial lumen in order to offer a scar point of attachmentor to prevent the backflow of drug in previous studies [12–16]. However, fibrin glue or bone cement is difficult tocontrol in the operating process. One animal died fromsuffocation caused by bone cement backflow into the tra-chea in Dong�s study [14] and the same accident occurred inone animal in our preliminary study as well. Therefore, it isunsafe to occlude bronchial lumens with bone cement.

Obvious marginal emphysema happened in two rabbits inthe bone cement occlusion group. Obstructive emphysemanot only influences the effectiveness of lung volumereduction but also increases the probability of pneumatoceleor aerothorax in clinical application. Twenty-three patientswith emphysema and bronchopleural fistula were treatedwith endoscopic Watanabe spigots during 2000 and 2001[20]. Complications reported were pneumonia (n = 2) anddyspnea (n = 1). In this study, obstructive pneumonia oc-curred in the bone cement occlusion group. Therefore, anyocclusion to bronchial lumen of the target area is notappropriate after drug infusion.

Our study develops a brand new approach for minimallyinvasive lung volume reduction in which alcohol and lipi-odol suspension is merely infused into the target areawithout occlusion to bronchial lumen. The merit of occlu-sion to bronchial lumens is to prevent drug backflow, butthis has already been avoided by our improved technique.

Summary

Alcohol and lipiodol suspension infusion together withballoon-tipped catheter occlusion can result in effectivelung volume reduction and has potential to be put intoclinical use. There are no other complications at presentother than a small quantity of backflow. Unlike broncho-scopic lung volume reduction [12, 13], no sterile abscessdeveloped with interventional lung volume reduction usedin this study.

Fig. 6. H&E stain of the target area of rabbit 9 shows thatpartial alveolus had interstitial pneumonia in the groupreceiving lipiodol alone (·100).

W. Wang et al.: Feasibility of ILVR 181

References1. Cooper JD, Patterson GS, Sundaresan RS, et al. (1996) Results of

150 consecutive bilateral lung volume reduction procedures in pa-tients with severe emphysema. J Thorac Cardiovasc Surg 112:1319–1329

2. Benditt JO, Lewis S, Wood DE, et al. (1997) Lung volume reductionsurgery improves maximal O2 consumption, maximal minute ventila-tion, O2 pulse, and dead space to tidal volume ratio during leg erg-ometry. Am J Respir Crit Care Med 156:561–566

3. Moy ML, Ingenito EP, Mentzer SJ, et al. (1999) Health-related qualityof life improves following pulmonary rehabilitation and lung volumereduction surgery. Chest 115:383–389

4. Criner GJ, Cordova SC, Furukawa S, et al. (1999) Prospective ran-domized trial comparing bilateral lung volume reduction surgery topulmonary rehabilitation in severe chronic obstructive pulmonary dis-ease. Am J Respir Crit Care Med 160:2021–2027

5. Geddes D, Davies M, Koyama H, et al. (2000) Effects of lung volumereduction in patients with severe emphysema. N Engl J Med 343:239–245

6. Goodnight-White S, Jones WJ, Baaklini J, et al. (2000) Prospectiverandomized controlled trial comparing bilateral lung volume reductionsurgery (LVRS) to medical therapy alone in patients with emphysema.Chest 118:102

7. Pompeo E, Marino M, Nofroni J, et al. (2000) Reduction pneumoplastyversus respiratory rehabilitation alone in severe emphysema: a ran-domized study. Ann Thorac Surg 70:948–954

8. Snell G, Holworth L, Borrill ZL, et al. (2003) The potential for bron-choscopic lung volume reduction using bronchial prosthesis: a pilotstudy. Chest 124:1073–1080

9. Toma TP, Hopkinson NS, Hillier J, et al. (2003) Bronchoscopic volumereduction with valve implants in patients with severe emphysema.Lancet 361:931–933

10. Yim AP, Hwong TM, Lee TW, et al. (2004) Early results of endoscopiclung volume redution for emphysema. J Thorac Cardiovasc Surg127:1564–1573

11. Gonzalez X, Dillard DH, Devore LJ, et al. (2002) Evaluation ofbronchoscopic and surgical lung volume reduction as single or com-bined procedures. Chest 122(Suppl):192–193

12. Ingenito EP, Reilly JJ, Mentzer SJ, et al. (2001) Bronchoscopic volumereduction: A safe and effective alternative to surgical therapy foremphysema. Am J Respir Crit Care Med 164:295–301

13. Ingenito EP, Berger RL, Henderson AC, et al. (2003) Bronchoscopicvolume reduction using tissue engineering principles. Am J Respir CritCare Med 167:771–778

14. Dong YH, Dong WH, Li HM, et al. (2004) Lung volume reduction bybronchial embolization in an animal model:a prelimimary report. AcadJ Sec Mil Med Univ 25:985–988

15. Ke W, Gan WC, You YF, et al. (2005) Pathological observation offunctional pulmonary lobectomy in canines. Chin J Med ImagingTechnol 21:183–186

16. You YF, Gan WC, Yin WH, et al. (2005) Animal model establishmentof functional pulmonary lobectomy. Chin J Intervent Imaging Ther2:58–62

17. Chung MP, Monick MM, Hamzeh NY, et al. (2003) Role of repeatedlung injury and genetic background in bleomycin-induced fibrosis. AmJ Respir Cell Mol Biol 29:375–380

18. Szapiel SV, Elson NA, Fulmer JD (1979) Bleomycin-induced interstialpulmonary disease in the nude,anthymic mouse. Am Rer Respir Dis120:893

19. Wang WS, Yao J, Wang WQ, et al. (2005) Experimental study on lungvolume reduction by bronchial embolization with lipiodol emulsion.Acta Univ Med Anhui 40:528–531

20. Toma TP, Matsuo K, Tamaoki A, et al. (2002) Endoscopic bronchialocclusion with spigots in patients with emphysema. Am J Respir CritCare Med 165(Suppl.):B9

182 W. Wang et al.: Feasibility of ILVR