Lung volume reduction surgery in canine model of predominantly upper lobe emphysema : advantages of new Surgical System – laboratory & animal investigations
Steven N. Mink
Objective: Lung volume reduction surgery has been shown to be an effective treatment for selected patients with advanced emphysema. Nevertheless, prolonged air leaks are a significant complication that limits the utility of this procedure. This study evaluated the safety and effectiveness of a novel surgical system designed to minimize this complication.
Methods: In 14 dogs, severe upper lobe emphysema was produced by repeated bronchial instillations of papain administered over an approximate 6-month interval. Pulmonary function testing that included lung volumes and flows was performed at baseline, after emphysema, and at 1 month and 6 months after resection in the surgical group, while at comparable intervals in the nonsurgical group. Seven animals were randomly assigned to a surgical group to test a vacuum-assisted surgical system (VALR Surgical System; Spiration; Redmond, WA) that deploys a compression silicone sleeve over portions of the diseased tissue. The other seven dogs comprised the nonsurgical group.
Results: In both groups, emphysema increased total lung capacity (TLC) approximately 125% as compared to baseline. In the surgical group, no air leaks were observed after resection, and TLC significantly decreased at the 1-month and 6-month periods as compared with postemphysema measurements. At necropsy, histologic examination revealed fibrosis of the compressed lung contained within the sleeve and fibrotic encapsulation of the device. Two animals had evidence of localized infection.
Conclusion: We successfully created a model of predominantly upper lobe emphysema. The vacuum-assisted surgical system provided safe and effective lung reduction without air leak complications and with sustained improvement in pulmonary function over 6 months.
Key words: air leak; dog; hyperinflation; lung volume reduction surgery; maximal expiratory flow; papain emphysema; pulmonary, emphysema; pulmonary function tests
Abbreviations: CL = tidal lung compliance; FE[V..sub.1]% = FE[V.sub.1] expressed relative to FVC;FRC = functional residual capacity; IC = inspiratory capacity; LVRS = lung volume reduction surgery; MEFV = maximum expiratory flow-volume; NETT = National Emphysema Treatment Trial; Pao = pressure at the airway opening; Ppl = pleural surface pressure; Ptp = transnpulmonary pressure P-V = deflation pressure volume; RL = airway resistance; RV = residual volume; TLC = total lung lung capacity; [V.sub.50] = maximal expiratory flow at 50% vital capacity
**********
Lung volume reduction surgery (LVRS) has been shown to be an effective treatment for selected patients with advanced emphysema. (1,2) During the operation, 20 to 35% of the most severely involved emphysematous lung is resected by means of either a median sternotomy or video-assisted thoracoscopy. LVRS is commonly performed with staple-cutting tools, with or without reinforcements of the staple lines with biocompatible materials or chemical adhesives. (1,2) Numerous controlled trials (1-5) have shown that volume reduction for emphysema can alter lung function, increase walking distance, and improve quality of life in selected patients. However, from the many studies, it is also clear that not all patients benefit from LVRS. (6,7) In the largest clinical trial, the National Emphysema Treatment Trial (NETT) (7) indicated that patients who had a very low FE[V.sub.1] (< 20% predicted), and either homogenous emphysema or a very low carbon monoxide diffusing capacity, had a high risk of dying from surgery. In this subgroup of patients from the NETT, most of the deaths occurred within 30 days of surgery and were related to respiratory failure, and most of the patients who died were receiving mechanical ventilation.
Air leaks are a frequent complication of LVRS that may contribute to prolonged mechanical ventilation and to the respiratory complications observed in these patients. (1) Better technical advances in LVRS, particularly those that reduce tissue manipulation and air leaks, may offer an improvement in the operative and postoperative mortality. In the present study, we developed a canine model of predominantly upper lobe emphysema to study the effect of LVRS on pulmonary function. The upper lobe model was developed since this distribution of emphysema has been shown to produce the most favorable response to LVRS in clinical trials. (2,6) To perform LVRS, we used prototypes of a novel vacuum-assisted lung tissue capture and reinforcement system (VALR Surgical System; Spiration; Redmond, WA) that has been designed to minimize air leaks. This system consists of an implantable compression silicone sleeve loaded into a cylindrical tool (introducer) that utilizes vacuum to draw the targeted tissue into the sleeve. Once deployed, the radial traction of the silicone sleeve compresses lung tissue producing volume reduction and allows tissue resection with minimal air leak complications (see “Materials and Methods” section).
We examined pulmonary function in this model at intervals of approximately 1 month and 6 months after LVRS (surgical group). We compared these results to animals with a similar degree of emphysema that did not undergo surgery (nonsurgical group). The objective of the study was to determine the physiologic effects, complications, air leak results, and lung pathologic changes that occurred over the 6-month interval when the VALR Surgical System was used to produce lung reduction in this canine model.
MATERIALS AND METHODS
Overall Design
These experiments were approved by the University of Manitoba Animal Care Committee and conform with the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health. (8) Of the 14 dogs enrolled in this study, 7 dogs were randomized at the beginning of the study to the surgical group, while the 7 other dogs comprised the nonsurgical group. Any animal that died during the period of emphysema induction was replaced into the appropriate group. After baseline pulmonary function testing was performed, severe heterogeneous emphysema was produced in all animals and took approximately 6 months to complete. LVRS was then performed in the surgical group with the VALR Surgical System. Pulmonary function test measurements were obtained at baseline, after emphysema was produced (postemphysema study), and at 1 month and 6 months after surgery. Measurements were obtained at equivalent intervals in the nonsurgical group.
Emphysema Model
A heterogeneous upper lobe distribution of emphysema was produced by multiple intrabronchial administrations of the enzyme papain. (9) The goal was to increase total lung capacity (TLC) to approximately 120% of baseline, and the number of instillations chosen was based on previous experience with this model. (9-11) In general, six instillations were placed into each of the upper lobes and one instillation into each of the lower lobes. (9) If TLC did not reach this value after the prescribed number of instillations as determined by pulmonary function measurements, then additional instillations (approximately two to three) were placed into the targeted lobes. This occurred in two animals in the surgery group and one animal in the nonsurgery group.
During the procedure, the dogs (23 to 30 kg) were anesthetized with sodium pentobarbital (.30 mg/kg), and an endotracheal tube was advanced into the trachea. (9-11) Under bronchoscopic visualization, 2.8 mL of the enzyme papain (Sigma Chemical; St. Louis, MO) mixed in 25 mL of sterile normal saline solution was instilled into one of the targeted lobes during the procedure. Instillations were repeated every other week, and the sequence was altered among the lobes of interest. After each instillation, the position of the animal was changed, so that the instilled lobe was placed dependently. Mechanical ventilation at 20 mL/kg (Harvard Apparatus; Holliston, MA) was performed for approximately 6 h in which the ventilator was connected to a source of 50% oxygen. Given in this dose and manner, the papain mixture spreads diffusely throughout the lobe of interest, but does not compromise oxygenation by spreading to the other lobes. All animals received antibiotics (clindamycin, 4 mg/kg) and gentamicin (2 mg/kg) before each instillation and every 12 h thereafter for a total of three doses to prevent bacterial infection.
Pulmonary Function Measurements.
Lung mechanics and maximum expiratory flow were determined at each study condition. The animals were anesthetized with pentobarbital sodium, 30 mg/kg. A large-bore armored endotracheal tube (inner diameter, 10 mm; Willy Rusch AG; Karman, Germany) was advanced into the trachea, and the lungs were mechanically ventilated (20 mL/kg). The animals breathed room air mixed with 5 L of oxygen that yielded an inspiratory oxygen concentration of approximately 35%. The esophageal balloon technique was used to measure pleural surface pressure (Ppl) as previously described, (11,12) in which the distal end of the 70-cm-long catheter attached to the esophageal balloon was connected to one port of a differential pressure transducer (Validyne Engineering; Northridge, CA). Pressure at the airway opening (Pao) was determined by means of a catheter attached to a port placed into the endotracheal tube. By means of two three-way stopcocks mounted on the respective ports of the differential pressure transducer, either Ppl, Pao, or transpulmonary pressure (Ptp) [Ptp = Pao – Ppl] were measured.
Determinations of TLC, functional residual capacity (FRC), and residual volume (RV) were made with the animal positioned in the left lateral decubitus position in a pressure-compensated volume displacement plethysmograph (Fig 1). (10-12) Lung volume was measured by a Krogh spirometer attached to the rear of the plethysmograph, while flow was measured by a pneumotachograph placed in series with the Krogh spirometer (Model DP; J. H. Emerson Company; Cambridge, MA).
FRC was determined by the plethysmographic method of Dubois et al, (13) in which the animal breathed against a valve that was closed at end-expiration. FRC was calculated from [([DELTA]V/ [DELTA]Pao) (PB- 47)], where PB is barometric pressure, and [DELTA]V/ [DELTA]Pao is the ratio of the change in thoracic gas volume to the change in Pao recorded on an oscilloscope and a chart recorder (Astro-Med; West Warwick, RI). (10-12) Following the FRC measurements, the lungs were deflated by means of a 1-L syringe to a Ptp of -10 cm [H.sub.2]O, and the gas volume removed defined expiratory reserve volume. RV was calculated as FRC minus expiratory reserve volume. The lungs were then inflated by a positive pressure source to Ptp = 30 cm [H.sub.2]O, and this volume defined inspiratory capacity (IC). TLC was calculated as IC + FRC.
After a standard volume history to TLC (Ptp = 30 cm [H.sub.2]O), deflation pressure-volume (P-V) curves were recorded on the oscillograph and were analyzed at 5 cm [H.sub.2]O Ptp decrements from TLC to end-expiratory lung volume. Tidal lung compliance (CL) was then determined from the change in Ptp measured between end-expiration and end-inspiration divided by tidal volume delivered by the ventilator.
Maximum expiratory flow volume (MEFV) and volume-time curves were performed by means of a negative pressure source that was attached to the airway opening. (10-12) After a TLC inflation, maximal flow and Ptp were recorded as a function of volume on the storage oscilloscope to obtain MEFV curves, while signals for flow, volume, and pressure were recorded as a function of time on the oscillograph to obtain expiratory volume vs time curves. FE[V.sub.1] was expressed relative to FVC (FE[V.sub.1]%). (12) From the oscillograph tracings, maximal expiratory flow at 50% vital capacity ([V.sub.50]) was obtained. (10,12) Airway resistance (RL) at 4 Hz was measured at a flow rate of 1 L/s, (12,14) by the subtraction technique of Mead and Whittenberger (14) in which the oscillatory component of Ptp in phase with flow divided by flow-determined RL.
LVRS With the VALR Surgical System
Surgery was performed at approximately 2 weeks after the postemphysema pulmonary function measurements were completed. Each animal was anesthetized with IV propofol and sufentanil. (15) The loading doses were 7.5 mg/kg and 1.25 [micro]g/kg, respectively, while the corresponding infusion rates were 0.18 mg/kg/min and 1 [micro]g/kg/min. A 9-mm or 10-mm endotracheal tube was positioned into the trachea, and the lungs were mechanically ventilated with 100% oxygen as the inspirate. An analgesic (buprenorphine, 0.05 mg/kg) was administrated subcutaneously. Antibiotics (clindamycin, 4 mg/kg, and gentamycin, 2 mg/kg) were administered through a catheter placed percutaneously into the femoral vein, while a percutaneously placed femoral artery catheter was used for determinations of arterial BP and blood gas analysis.
The animal was placed into the supine position. Under sterile conditions, a midline incision was made to expose the thoracic organs. Lung units that were emphysematous could easily be identified by their yellowish, hyperexpanded, and transparent appearance as compared to normal lung units. Severe emphysema was mostly observed in the lapper lobes (see “‘Results” section).
Standard techniques were used to perform LVRS. An endobronchial blocker (Cook Veterinary Products; Bloomington, IN) was placed at the origin of the right bronchus to isolate the right lung. Following bronchial occlusion, the emphysematous portion of the lung required 15 to 20 min for incomplete deflation as compared with the 2 to 5 min required for the normal-appearing lung. The prototype VALR Surgical System provided for this study consisted of a thin silicone sleeve (compression sleeve) loaded inside a polycarbonate cylinder introducer. The proximal end of the sleeve contains architecture designed for gripping onto the lung tissue and was everted onto the proximal end of the introducer before its use (Fig 2). The introducer was connected to a regulated vacuum source that allowed for a nontraumatic, gentle capture of selected lung tissue into the sleeve (Fig 3). When the controlled vacuum was activated, this pressure did not exceed 8 inches Hg. Once the lung was captured, the sleeve was released from the introducer by advancement of the outer cylinder on the introducer. The compression sleeve produced radial compression and prevented re-expansion of the targeted lung tissue.
[FIGURES 2-3 OMITTED]
The proximal section of the silicone sleeve was then secured and affixed to the lung tissue by means of a U-stitch polyester 0 surgical suture, and another U-stitch suture was placed at a distance of approximately 1 inch from the proximal suture. From this point, most of the sleeve and reduced tissue was resected, such that only the proximal sutured portion of the sleeve remained on the reduced lobe (Figs 3, 4). Only one silicone sleeve was required for each of the right upper lobes to achieve significant reduction of the diseased lobe. When work on the right lung was completed, the endobronchial blocker was deflated and the right lung was expanded. An identical procedure was performed for the left upper lobe in which two sleeves usually were required (see “Results” section).
[FIGURE 4 OMITTED]
When all implants were completed, careful observations were made for air leaks by filling the thoracic cavity with warm 0.9% saline solution and looking for bubbles during inflation and mechanical ventilation. After the saline solution was removed, a chest tube was placed in the right hemithorax and the chest incision was closed in planes.
The chest tube was connected to a Heimlich valve, and the valve was connected to a vacuum water-trap system. Air leaks were evaluated for approximately 30 to 200 min, and the presence or absence of air leaks was documented. When no bubbles were observed, the water-trap system was disconnected, and the chest tube was removed. The animals remained on oxygen support until they maintained adequate oxygenation and ventilation on their own. When considered stable, the animal was extubated and returned to animal housing where daily observations and administration of analgesics and antibiotics were performed for 1 week.
Euthanasia and Necropsy
Approximately 3 weeks alter the 6-month pulmonary function measurements were completed, the animals were anesthetized with pentobarbital, 30 mg/kg. An endotrachheal tube was advanced into the trachea, and the lungs were mechanically ventilated while the animals breathed 100% oxygen. Under sterile conditions, the chest was opened via median sternotomy. Observations were made on tissue and implant conditions, adhesions, fibrosis, and presence of fluid in the thoracic cavity. Cultures were obtained for microbiology from implant and control sites. After observations were completed, each animal was terminated with an IV injection of euthanasia solution (Euthasol; Delmarva Laboratories; Midlothian, VA). The lungs were returned en bloc, the trachea was cannulated, and the lung tissue was infused with 10% formalin solution that was maintained at 20 to 25 cm [H.sub.2]O pressure for 24 h for tissue fixation, after which sectioning for histology was performed.
Histology
A board-certified American College of Veterinary Pathologists physician who functioned independently of the study investigators evaluated the tissue sections. The tissue from the surgical and nonsurgical groups was sectioned in slices of 0.2- to 0.4-cm thickness and embedded in paraffin. The slides were stained with hematoxylin-eosin by standard methods. In the nonsurgical group, the primary parameters evaluated were the degrees of emphysema, inflammation, and fibrosis. In the surgical group, the reduced tissue and distal nonreduced tissue were also examined. Grading of most parameters utilized a 0 to 4 scale, where 0 = none, 1 minimal, 2 mild, 3 = moderate, and 4 = severe/complete. Presence of bacterial colonies was also noted on the histologic samples.
Statistics
Statistical analyses included a two-way (between-within), and a three-way repeated measures (between-within-within) analysis of variance. A Student Newman-Keuls multiple comparison test was used to determine where differences were found among the various conditions in the two groups. Results are reported as mean [+ or -] 1 SD.
RESULTS
Seventeen animals were initially enrolled into the study. Three animals died before the emphysema treatments were completed. Two animals died due to complications related to the immediate effects of a papain instillation that resulted in acute respiratory failure, while one of the animals was killed because of the repetitive development of seizures when anesthesia was administered. In addition, two other animals (one in the surgical group and one in the nonsurgical group) acquired a pneumothorax over the course of papain administration that was related to the pulmonary function measurements; it was successfully treated by chest tube drainage. Of the 14 dogs that completed the protocol, 7 dogs were enrolled in the surgical LVRS group and 7 dogs were in the nonsurgical group. All were active and healthy for the duration of the study.
Pulmonary Function Measurements
At baseline (before emphysema), TLC was nearly identical in both the surgical and nonsurgical groups and measured 3.45 L and 3.43 L, respectively (Fig 5, top). After emphysema, TLC increased to approximately 125% of baseline in both groups. In the surgical group, volume reduction at the 1-month measurement was associated with a return in TLC to slightly below the baseline value. At the 6-month measurement, TLC increased slightly (0.4 L), as compared with the 1-month study, but was still 0.6 L below the postemphysema measurement. In the nonsurgical group, TLC remained unchanged at the 1-month and 6-month studies as compared with the postemphysema study.
The changes in FRC and RV followed those in TLC (Fig 5, middle, bottom). In both the surgical and nonsurgical groups, FRC increased approximately 150% after emphysema as compared with baseline. In the surgical group, volume reduction at the 1-month period was associated with a return in FRC to the baseline value, while there was little change in FRC between the 1-month and 6-month studies. In the nonsurgical group, FRC remained unchanged at the 1-month and 6-month studies as compared with the postemphysema study. In both the surgical and nonsurgical groups, RV increased to a similar extent in both groups after emphysema was produced. In the nonsurgical group, RV remained unchanged at the 1-month and 6-month periods as compared with the postemphysema study. In the surgical group, volume reduction at the 1-month and 6-month studies was associated with a decrease in RV as compared with the postemphysema study, but RV was not significantly different between the two groups.
[FIGURE 5 OMITTED]
The deflation P-V curves for the surgical and nonsurgical groups are shown in Figure 6. In the nonsurgical group, the P-V curve after emphysema was shifted upward as compared with the baseline, and the P-V curve remained unchanged as compared with the postemphysema study at the 1-month and 6-month studies. In the surgical group, the P-V curve showed a similar upward shift at the postemphysema study, and returned to the baseline curve at the 1-month study. There was a slight upward shift in the curve at the 6-month study as compared with the 1-month study. CL and RL did not change after emphysema was produced, and were not different between the two groups over the course of the study (Table 1).
[FIGURE 6 OMITTED]
Emphysema was associated with marked changes in maximal expiratory flow. After emphysema, there was a marked decrease in flow rates as compared with baseline (Fig 7, Table 2). In the surgical group, Vso and FE[V.sub.1]% increased between the postemphysema study and either the 1-month or 6-month studies, and these increases were significantly greater than those measured in the nonsurgieal group.
[FIGURE 7 OMITTED]
Lung Volume Reduction With the VALR Surgical System
During surgery, all animals had evidence of severe heterogeneous emphysema in which the upper lobes were hyperinflated, bullous, and had extensive air trapping. The less emphysematous lobes were compressed and restricted by the upper lobe hyperinflation and, in some areas, segmental atelectasis was observed. In the surgery group, implantation of only one silicone sleeve in the right upper lobe was sufficient to reduce 50 to 75% of the lobe mass. The sleeve size diameter used for this lobe had the following distribution: 10 mm (n = 3), 12 mm (n = 3), and 15 mm (n = 1). In the left upper lobe, two devices were typically deployed in each animal. One device was placed in the lobe apex with the following diameter size distribution: 8 mm (n = 1), 10 mm (n = 4), and 12 mm (n = 2) to reduce 25 to 50% of the lobe mass. Another device was placed in the anterior medial aspect: 8 mm (n = 1), 10 mm (n = 4), and 12 mm (n = 1) diameter to reduce 25 to 50% of the lobe mass.
After lung volume reduction was completed, evaluation of perioperative air leaks was performed. With the chest filled with warm saline, no air leaks were observed from all the implant sites or neighboring nonreduced lung tissue. Evaluation of postoperative air leaks was subsequently determined by attachment of a chest tube to a Heimlich valve connected to a water-trap system. After thorax evacuation, no air leaks were observed in six animals of the surgical group; in only one animal, a small intermittent leak was observed. This bubbling stopped prior to chest tube removal. In all dogs, chest tubes were removed 1 to 4 h after surgery. All dogs recovered without problems from surgery and were active in their cages the following day. All animals were in good condition at the 6-month evaluation after LVRS.
During the volume reduction procedure, the only complications observed were chest wall bleeding in one animal that required reoperation to repair a branch of the mammary vein, while in another animal wound dehiscence occurred approximately 3 days after surgery and two more times in the following weeks. This dehiscence appeared self-inflicted and required resuturing of the skin and chest wall muscles; unfortunately, the incision became infected and required wound care and prolonged antibiotic therapy.
Necropsy and Histology
At necropsy, animals enrolled in the nonsurgical group had evidence of moderate-to-severe emphysema with clearly visible dilated airspaces on gross inspection (Fig 8, top) in the upper lobes and mild emphysema in the lower lobes. In the surgical group, emphysema was evident in the nonreduced tissue. In most implant locations, devices were found in place covered by thin fibrotic encapsulation limited to the silicone material. The lung tissue identified inside the device appeared fibrotic, shrunken and contracted within the device (Fig 8, middle). Device migration, defined as separation and movement of the silicone band from the original placement location, was observed in two sites in one dog, but there were no clinical consequences as the device probably moved after tissue healing and fibrosis had occurred. Both sleeves were found attached and fibrosed to the midline incision of the thorax. In all reduced areas, there was no evidence of any leak when the lung was inflated with formalin. In most cases on visual inspection and in all cases on histologic examination, the reducing devices were found to be encapsulated by fibrotic tissue (Fig 8, bottom).
[FIGURE 8 OMITTED]
In the nonsurgical group, the histologic sections showed emphysematous changes throughout the lungs characterized by enlarged, coalesced irregular airspaces with destruction of septal walls. The upper lobes were often graded as 4 (most severe). The emphysematous areas were noted to compress the adjacent and intervening alveolar airspaces, and resulted in mild collapse with slit-like airspaces of the compressed areas. In addition to the emphysema, mild chronic inflammatory changes that included variable combinations of macrophages and mild pleural and parenchymal fibrosis were also observed. In the surgical group, the tissue subjected to the devices was well compressed and fibrotic, in which coagulative necrosis and chronic inflammation were the predominant findings present. Fibrotic encapsulation of the devices was complete with minimal involvement of the lung or pleura with fibrosis. Fibrosis and inflammation extended only to a minimal degree to the nonreduced lung tissue.
Two of the seven animals in the surgical group had evidence of limited bacterial infection in areas associated with the devices. One of these animals required resuturing of the chest wall muscles because of wound dehiscence. Wound infection in this animal may have contaminated the thorax and led to migration of the device. In the two animals with infection, the histologic samples showed attendant neutrophilic or pyogranulomatous inflammation in the areas of infection and microbiology revealed Pseudomonas aeruginosa. However, in the nonreduced tissue adjacent to the device, there was no evidence of bacterial infection present, so that the infection was limited to the areas where the device was implanted.
DISCUSSION
For this study, we developed a canine model of predominantly heterogeneous upper lobe emphysema to examine a novel method by which lung volume reduction could be achieved without the use of staples or cutting tools. This model may be superior to other emphysema models created to test lung volume reduction (16) because of the degree of pathology produced in the upper lobes and the heterogeneous distribution of the disease that has been shown to respond best to LVRS in clinical trials. (2) The increases in TLC, FRC, and RV found at the postemphysema study were consistent with the development of severe hyperinflation and air trapping. These lung function changes were similar to those found in patients with severe emphysema who would undergo LVRS in clinical medicine. (1-6) The results showed that the VALR Surgical System produced effective lung volume reduction in this model. At the 1-month evaluation, there was a return of TLC and FRC to baseline values. Although RV decreased in six of seven dogs at the 1-month study, it did not return to the baseline value. Since many areas of emphysematous lung were not removed during LVRS, the lack of a return of RV to the baseline value most likely indicates that a significant amount of nonresected emphysematous lung units still remained in the surgical group.
At the 6-month evaluation, there was a significant increase in TLC in the surgical group as compared with the 1-month period, although TLC was still significantly below that measured at the postemphysema study. In the nonsurgical group, there were no changes in lung function found between the 1-month and 6-month studies. The latter observation would indicate that the lung function changes of emphysema are relatively stable in this model once the lesion has been produced, and that the increase in TLC observed at the 6-month study in the surgical group could not be explained by further hyperinflation due to the development of more emphysema.
However, we think that the increase in TLC found in the surgical group at the 6-month period reflected the recruitment of lung units that were observed compressed by surrounding hyperinflated lung units during the surgical procedure. Of interest, it appears that the lung units recruited represented relatively healthy units rather than emphysematous ones for the following reason. As noted in the results, RV in the surgical group did not change between the 1-month and 6-month studies, while TLC increased approximately 0.4 L. This means that the slow vital capacity (TLC – RV) must have also increased to a similar extent. Not only did we find this to be the case (slow vital capacity increased from 2.7 to 3.1 L), but on the mean both FE[V.sub.1] and FVC also increased to a similar extent of 0.4 L between the 1-month and 6-month studies (Table 2). When we calculated the FE[V.sub.1]/FVC of these newly recruited units, the ratio was approximately 1.0 (0.4/0.4), which is the normal ratio found for healthy canine lung. However, if the increase in TLC were due to the recruitment of emphysematous rather than healthy lung units, then the FE[V.sub.1]/FVC ratio of these newly recruited units would have been much less than 1.0.
Furthermore, in terms of the pulmonary function protocol used, we standardized TLC by inflation of the lung to 30 cm [H.sub.2]O Ptp in the two groups. We inserted an esophageal balloon to measure pleural pressure and used the difference between airway opening pressure and pleural pressure to determine Ptp. (10-12) Since Ptp measured at TLC would be the same in the emphysematous and nonemphysematous lung units, lung volume in the two groups would be determined by the relative compliances of the different lung units, and not by factors such as compression of the different lobes or by changes in chest wall properties that might occur over the course of the study. We therefore do not think that there should be any bias in the interpretation of the measurements of the lung volumes between the surgery and nonsurgery groups.
In the surgical group, the results indicate that the VALR device produced effective lung volume reduction over the 6-month interval in this emphysema model. This system, which utilizes a novel technology of vacuum-assisted volume reduction, makes it easy to perform bilateral LVRS. This system allows tissue capture and gentle manipulation, and thereby reduces the likelihood of trauma on the delicate lung tissue. After device deployment, suture fixation and tissue resection, only a small portion of the device was left behind that banded and produced radial compression of the reduced tissue. These features may be a distinctive advantage over conventional stapling techniques.
We nevertheless recognize that our experience was limited to reduction of only the apical and anterior aspects of the upper lobes, and that we did not test whether access to other regions may be more difficult given the rigid and large size of the instrumentation used in the present study. It is evident however that the VALR Surgical System allowed deployment of one or two appropriately sized silicone sleeves per lobe replacing multiple staple firings, providing adequate volume reduction and a safe substrate for tissue resection. Due to the differences in human anatomy and morphometry, evaluation and adjustments on the sleeve sizes may be required before this system is considered for human use. We further acknowledge that the duration of emphysema was relatively short term, and the emphysematous lesion may not be as severe as that found in the human condition. Human emphysematous lungs may also be compromised due to concurrent conditions related to previous infections or pneumoconiosis that may make it difficult to use this device and the presence of extensive adhesions at the time of surgery. Finally, although the procedure was easy to perform in the flexible small canine thorax, access in the human thorax may provide additional limitations.
Prolonged air leaks are a common complication of LVRS. Cooper et al (1) found prolonged air leaks of > 7 days in 46% of their 150 patients who undertook this procedure. Indeed, in the latter study, (1) some patients had chest tubes in place for up to 38 days postoperatively. Since prolonged air leaks would limit patient mobility, prolong hospital stay, and promote infection due to compression of lung tissue, this is an important procedural complication to be avoided. Many patients in the NETT were receiving mechanical ventilation due to respiratory failure at the time of death, (7) and the authors wonder the extent to which air leaks contributed to the respiratory failure observed in these patients.
The VALR Surgical System minimized problems with air leaks known to occur when the lung is resected by staples and cutting tools. In all of the seven animals that underwent LVRS in the present study, none of the dogs required prolonged chest tube drainage. We made careful observations for air leaks in the perioperative and postoperative periods, and found no evidence of air leaks in the surgical group. Similar findings of minimal air leaks were reported by Brenner et al, (16) who performed volume reduction surgery in a rabbit model of emphysema with a similar prototype system.
In the design of the VALR Surgical System, many safeguards were placed on the sleeve that would prevent its dislodgement, which included suturing the sleeve. In one animal, in which two devices were noticed to have migrated, no air leak was observed, even when the lung was subjected to high transpulmonary pressures during the infusion of formalin for fixation. These results would therefore suggest that even if the device migrated in the in vivo condition, the risk of air leak would be small after a period of time because the bronchi and parenchyma contained within the device would gradually become occluded by fibrotic tissue. Although no problems with air leaks were observed in the present study, we agree that only a small number of animals was studied, and the degree of emphysema may not have been as severe as that found in the human condition. Also, a minimal degree of adhesions was observed in this model, and this may be an important difference compared to LVRS in humans where pleural tearing of adhesions proximal or distal to the areas targeted for volume reduction frequently occurs during dissection, tissue manipulation, and lung inflation.
The consideration of infection is an important one in this study, since two animals had histologic and microbiological evidence of infection in the surgical group. In one dog, this was most likely related to wound dehiscence since interrupted sutures were used to close the sternum, and the animal ripped open the continuous sutures that were used to close the muscles and skin. In the other animal, no explanation for the infection could be discerned and may be related to contamination during surgery, bacteremia, or occurred by the inhaled route, although blood culture results were negative for bacteria. In both animals, the infection remained localized to the area of lung tissue that surrounded the small silicone band, and the animals remained healthy with no signs of systemic infection observed.
We think that an important aspect in the prevention of infection would be to keep the implantable sleeve as short as possible when it is inserted. Since the sleeve may act as a focus for infection, it is likely that minimizing the length of the sleeve would be advantageous in preventing this complication. In addition, since we now know that the lung tissue that is compressed by the device eventually becomes fibrotic and that the device itself becomes encapsulated by fibrotic tissue, a longer course of antibiotics and even antibiotic impregnated devices may eliminate the likelihood of infections. Conceivably, once the device became encapsulated, the fibrotic tissue could shield the device from bacteria that would otherwise come in contact with it either by the inhaled or bloodstream route. In that case, the chance of infection would be minimized with the VALR device.
In terms of the emphysema model described in the present study, by selective administration of the enzyme, we were able to produce a heterogeneous emphysema distribution with more severe effects in the upper lobes. In addition, this model is primarily one of severe hyperinflation, since the increase in airway resistance related to accompanying chronic bronchitis found in the clinical situation is minimal (Table 1). However, we recognize that the lesion produced is more characteristic of panacinar emphysema than centrilobular emphysema. (17) The former is found in patients with [[alpha].sub.1]-antitrypsin deficiency, while the latter is found in smokers. (18) Although this model is more applicable to chronic obstructive lung disease patients with emphysema alone, such patients are the most likely ones to benefit from the lung volume reduction procedure in clinical medicine.
In summary, a canine model of heterogeneous, predominantly upper lobe emphysema was successfully created to evaluate a new system to perform LVRS. The VALR Surgical System produced safe and effective treatment in the surgical group as compared to a control group. The procedure produced minimal morbidity and no air leak complications were observed. Although further experiments may need to be performed, these features may make the VALR Surgical System advantageous for reducing postoperative complications in patients who undergo LVRS.
[FIGURE 7 OMITTED]
Table 1–RL and CL in the Surgical and Nonsurgical Groups *
Surgical Group
Variables Baseline After Emphysema
RL, cm [H.sub.2]O/L/s 2.04 [+ or -] 0.36 2.64 [+ or -] 0.27
CL, mL/[H.sub.2]O 171 [+ or -] 73 140 [+ or -] 25
Surgical Group
Variables 1-mo Study 6-mo Study
RL, cm [H.sub.2]O/L/s 3.51 [+ or -] 2.89 2.23 [+ or -] 0.40
CL, mL/[H.sub.2]O 134 [+ or -] 43 195 [+ or -] 56
Nonsurgical Group
Variables Baseline After Emphysema
RL, cm [H.sub.2]O/L/s 2.09 [+ or -] 0.21 2.69 [+ or -] 0.49
CL, mL/[H.sub.2]O 154 [+ or -] 35 161 [+ or -] 63
Nonsurgical Group
Variables 1-mo Study 6-mo Study
RL, cm [H.sub.2]O/L/s 2.56 [+ or -] 0.27 2.50 [+ or -] 0.46
CL, mL/[H.sub.2]O 195 [+ or -] 70 150 [+ or -] 18
Table 2–Parameters of Maximal Expiratory Flow in the Surgical
and Nonsurgical Groups *
Surgical Group
Variables Baseline After Emphysema
[V.sub.50], L/s 6.1 [+ or -] .7 2.6 [+ or -] 1.9 ([dagger])FE[V.sub.1], % 98 [+ or -] 2 76 [+ or -] 12 ([dagger])
FE[V.sub.1], L/s 3.6 [+ or -] .2 2.6 [+ or -] .5 ([dagger])
FVC, L 3.65 [+ or -] .2 3.5 [+ or -] .4
Surgical Group
Variables 1-mo Study
[V.sub.50], L/s 4.3 [+ or -] 1.8 ([dagger]) ([section])FE[V.sub.1], % 86 [+ or -] 7 ([dagger]) ([section])
FE[V.sub.1], L/s 2.16 [+ or -] .4 ([dagger]) ([double dagger])
([section])
FVC, L 2.54 [+ or -] .6 ([dagger]) ([double dagger])
([section])
Surgical Group
Variables 6-mo Study
[V.sub.50], L/s 4.4 [+ or -] 1.7 ([dagger]) ([section])FE[V.sub.1], % 85 [+ or -] 6 ([dagger]) ([section])
FE[V.sub.1], L/s 2.5 [+ or -] .5 ([dagger])
FVC, L 2.94 [+ or -] .5 ([dagger]) ([section])
Nonsurgical Group
Variables Baseline After Emphysema
[V.sub.50], L/s 5.9 [+ or -] .9 3.8 [+ or -] 1.9 ([dagger])FE[V.sub.1], % 99 [+ or -] 1 84 [+ or -] 5 ([dagger])
FE[V.sub.1], L/s 3.5 [+ or -] .3 2.7 [+ or -] .8 ([dagger])
FVC, L 3.5 [+ or -] .4 3.17 [+ or -] .8
Nonsurgical Group
Variables 1-mo Study
[V.sub.50], L/s 3.8 [+ or -] 1.9 ([dagger])FE[V.sub.1], % 84 [+ or -] 6 ([dagger])
FE[V.sub.1], L/s 2.7 [+ or -] .6 ([dagger])
FVC, L 3.24 [+ or -] .6
Nonsurgical Group
Variables 6-mo Study
[V.sub.50], L/s 3.5 [+ or -] 1.5 ([dagger])FE[V.sub.1], % 88 [+ or -] 5 ([dagger])
FE[V.sub.1], L/s 2.8 [+ or -] .5 ([dagger])
FVC, L 3.23 [+ or -] .4
* Data are presented as mean [+ or -] SD.
([dagger]) p < 0.05 vs baseline, two-way analysis of variance
and Student Newman-Keuls.
([double dagger]) p < 0.05 vs nonsurgical group.
([section]) p < 0.05, significantly different change in parameter
between 6-month study or 1-month study and postemphysema study in
the surgical vs non-surgical group.
ACKNOWLEDGMENT: The authors thank Dr. Piper Treuting, DVM, Board Certified American College of Veterinary Pathologist, for histologic analysis.
REFERENCES
(1) Cooper JD, Patterson GA, Sundaresan RS, et al. Results of 150 consecutive bilateral lung volume reduction procedures in patients with severe emphysema. J Thorac Cardiovasc Surg 1996; 112:1319-1330
(2) National Emphysema Treatment Trial Research group. A randomized trial comparing lung-volume reduction surgery with medical therapy for severe emphysema. N Engl J Med 2003; 348:2059-2072
(3) Geddes D, Davies M, Kayama H, et al. Effect of lung volume reduction surgery inpatients with severe emphysema. N Engl J Med 2000; 343:239-245
(4) Criner GJ, Cardova FC, Furkawa S, et al. Prospective randomized trial comparing bilateral lung volume reduction surgery to pulmonary rehabilitation in severe chronic obstructive pulmonary disease. Am J Respir Crit Care Med 1999; 160:2018-2027
(5) Pompeo E, Marino M, Nofroni I, et al. Reduction pneumoplasty versus respiratory rehabilitation in severe emphysema: a randomized study; Pulmonary Emphysema Research Group. Ann Thorac Surg 2000; 2000; 70:948-953
(6) MeKenna RJ, Brenner M, Fischel RJ, et al. Patient selection criteria for lung volume reduction surgery. J Thorac Cardiovasc Surg 1997; 14:957-967
(7) National Emphysema Treatment Trial Research group. Patients at high risk of death after lung volume-reduction surgery. N Engl J Med 2001; 345:1075-1083
(8) Guide for the care and use of laboratory animals. Bethesda, MD: National Institutes of Health, 1996; publication No. 85-23
(9) Mink SN. Mechanism of lobar alveolar pressure decline during forced deflation in canine regional emphysema. J Appl Physiol 1997; 82;632-643
(10) Mink SN. Expiratory flow limitation and the response to breathing a helium-oxygen gas mixture in a canine model of pulmonary emphysema. J Clin Invest 1984; 73:1321-1334
(11) Mink S, Gomez A, Whitley L, et al. Hemodynamics in dogs with pulmonary hypertension due to emphysema. Lung 1986; 164:41-54
(12) Mink S, Coalson JJ, Whitley L, et al. Pulmonary function tests in the detection of small airways obstruction in a canine model of bronchiolitis obliterans. Am Rev Respir Dis 1984; 130:1125-1133
(13) DuBois ASY, Botelho G, Bedell N, et al. A rapid plethysmographic method for measuring thoracic gas volume: a comparison with nitrogen washout method for measuring functional residual capacity in normal subjects. J Clin Invest 1956; 35:322-326
(14) Mead J, Whittenberger JC. Physical properties of human lung measured during spontaneous respiration. J Appl Physiol 1953; 5:779-796
(15) Flecknell P. Laboratory anesthesia. 2nd ed. San Diego, CA: Academic Press, 1996; 12
(16) Brenner M, Gonzales X, Jones B, et al. Effects of a novel implantable elastomer device for lung volume reduction surgery in a rabbit model of elastase-induced emphysema. Chest 2002; 121:201-209
(17) Pusphakom R, Hogg JC, Woolcock AF, et al. Experimental papain-induced emphysema in dogs. Am Rev Respir Dis 1979; 102:778-789
(18) Fishman AF. Chronic obstructive lung disease: overview. In: Fishman AP, Elias JA, Fishman JA, et al., eds. Fishman’s pulmonary diseases and disorders. 3rd ed. New York, NY: McGraw-Hill, 1998; 645-658
* From the Section of Respiratory (Drs. Mink and Bautista, and Ms. Duke), Department of Internal Medicine, and Section of Thoracic Surgery (Dr. Tan), Department of Surgery, University of Manitoba, Winnipeg, MB; and Spiration Inc. (Dr. Gonzalez), Redmond, WA.
This study was supported in part by a grant from Spiration, Inc., and in part by a grant from the Manitoba Medical Services Foundation.
Dr. Bautista was supported by a Fellowship from the Manitoba Lung Association.
Manuscript received March 6, 2003; revision accepted July 10, 2003.
Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (e-mail: permissions@ehestnet.org).
Correspondence to: Steven N. Mink, MD, FCCP, GF-221, Health Sciences Centre, 700 William Ave, Winnipeg, MB R3E-0Z3, Canada; e-mail: minksn@cc.umanitoba.ca
COPYRIGHT 2004 American College of Chest Physicians
COPYRIGHT 2004 Gale Group
