SEARCH

SEARCH BY CITATION

ABSTRACT

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIAL AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. DISCLOSURE
  8. REFERENCES

Objective

To describe a surgical technique for thoracoscopy and report visible anatomy within the thoracic cavity of standing cattle.

Study design

Prospective study.

Animals

Adult clinically healthy Holstein–Friesian cows (n = 15).

Methods

Each cow had four thoracoscopic examinations. Initially, the left hemithorax was examined after passive lung collapse, then again 24 hours later after CO2 insufflation. The right hemithorax was examined 24 hours later after passive lung collapse and again 24 hours later after CO2 insufflation.

Results

CO2 insufflation did not significantly improve visibility within the pleural space. Collapsed lung, aorta, esophagus, diaphragm, and azygos vein were readily viewed; however, the pericardial region was not consistently visible. Minor laceration of the lung occurred in 1 cow with adhesions, otherwise there were no intra- or postoperative complications. All cows recovered without signs of discomfort. No local swelling or emphysema occurred at the portals.

Conclusions

Thoracoscopy can be safely performed on healthy standing cattle.


INTRODUCTION

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIAL AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. DISCLOSURE
  8. REFERENCES

Thoracoscopy is a minimally invasive procedure to examine the pleural space with rigid or flexible endoscopes,[1] and can complement conventional diagnostic methods (clinical examination, ultrasonography, radiography, and endoscopy). The role of imaging procedures is becoming increasingly important in bovine medicine.[2] We are only aware of a single case report in cattle where thoracoscopy was used as an additional technique for diagnosis and treatment of pericardial lymphoma.[3]

Our purpose was to describe a thoracoscopic technique in standing cattle with and without CO2 insufflation, report visible structures and potential operative and postoperative complications. We hypothesized that (1) thoracoscopy can be safely performed in standing cattle; (2) most structures within the thoracic cavity would be visible; and (3) intrapleural CO2 insufflation would significantly improve observation within the pleural space.

MATERIAL AND METHODS

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIAL AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. DISCLOSURE
  8. REFERENCES

The study was approved by the Animal Welfare Commission (TVV-No. 14/03) and was part of a larger study on assessment of cardiovascular and respiratory responses to thoracoscopy in healthy standing cattle.

Cattle

Healthy adult female Holstein–Friesian cattle (n = 15; aged 3–9 years; mean, 5 years) with a mean weight of 483 kg were studied. Clinical examination and blood analysis (hematology, serum biochemical profile, and blood gases) were used to assess that the cows were healthy. Cows were housed in individual boxes.

Study Design

Each cow had 4 thoracoscopic examinations. Initially, the left hemithorax was examined after passive lung collapse, and then 24 hours later the left hemithorax was examined after CO2 insufflation. Twenty-four hours later the right hemithorax was examined after passive lung collapse and then 24 hours later after CO2 insufflation. Five cows were examined in the 8th, 9th, or 10th intercostal space, respectively. The cows were selected randomly for the attribution of the intercostal spaces. A simple randomization method (drawing numbers) was used.

Preoperative Preparation

Cows had free access to food and water before surgery. For thoracoscopy, cows were restrained in standing stocks and the tail was tied to the halter on the contralateral side to the surgical procedure. The skin extending from the shoulder joint to the caudal angle of the scapula and caudally to the paralumbar fossa was clipped and aseptically prepared. A sterile drape fixed to the skin by adhesive tape was used to cover the thorax and abdomen. Portal sites were anesthetized with 2% procaine (15–20 mL) injected locally into the subcutaneous, muscular, and pleural tissues of the selected intercostal space.

Surgical Technique

The endoscopic portal in each interspace was aligned with the ventral margin of the tuber coxa. A #10 scalpel blade was used to create a 1.5-cm vertical stab incision through the skin and subcutaneous tissues in the intercostal space, being careful to avoid the neurovascular bundle on the caudal aspect of the cranial rib. A 9-cm long blunt stainless teat cannula (Hauptner & Herberholz GmbH & Co. KG, Solingen, Germany) was carefully introduced into the pleural space to create pneumothorax. After 3–4 respiratory cycles, the teat cannula was removed and replaced by a 20-cm long, 11-mm diameter cannula with a sharp guarded trocar (Dr. Fritz GmbH, Tuttlingen, Germany), which was introduced perpendicular to the rib cage using a rotatory motion. After entry, the trocar was removed and a 30°, 57-cm long, 10-mm diameter rigid telescope (Storz GmbH & Co. KG, Tuttlingen, Germany) was inserted through the cannula. A camera (Tricam SL; Storz) and a xenon light source cable (Xenon 300 and Fluidcable; Storz) was attached to the endoscope.

During thoracoscopy with passive lung collapse, the stopcock of the cannula remained open. A suction unit (securat80-GF 200/GF 210; Aesculap, Tuttlingen, Germany) was connected to the side of the cannula to evacuate air or CO2 at the end of the procedure. For thoracoscopy under CO2 insufflation, a pressure-limited insufflating device (Eurotherm; Dr. Fritz GmbH) was connected to the cannula to maintain a constant pressure of 5 mmHg.

Thoracoscopy was performed for 20 minutes. Each examination followed a standardized pattern. Initially, the endoscope was moved in a dorsocranial direction, then dorsally, caudally, and finally in the ventral direction. At the end of each examination, air or CO2 was slowly evacuated by the suction unit to reestablish negative pressure in the pleural space. Complete lung re-inflation was verified as the endoscope was slowly withdrawn from the pleural space, then the endoscope and cannula were removed and the skin closed using 0 nylon in a simple interrupted pattern. Sterile gauze and an adhesive drape were used to protect the portal sites.

Examinations were recorded on videotape for later viewing. Digital images were stored, then edited (Adobe® Photoshop CS4 and Corel® PHOTO-PAINT X3).

Postoperative Evaluation

After surgery, cows were returned to their stall. Cows were monitored for behavioral signs of pain or discomfort, respiratory distress, and for portal site heat, pain, or emphysema, and for the first 9 days also had a physical examination every 12 hours. Twenty-four hours after the last thoracoscopic intervention, laterolateral radiographic projections of the thorax were used to confirm the absence of a pneumothorax.

Sutures were removed at 10 days and cows were discharged for slaughter.

Statistical Analysis

Descriptive statistics were calculated and the Shapiro-Wilk test was used to analyze data for normality. Statistical analysis of within-subject groups was performed using McNemar's χ2 test. Analyses were performed with software (SPSS15, SPSS Software GmbH, Munich, Germany) and significance was set at P < .05.

RESULTS

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIAL AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. DISCLOSURE
  8. REFERENCES

Thoracoscopy was tolerated with minimal restraint and successfully completed in all cows.

The thoracoscopic technique described allowed thorough observation of the pleural space. Thoracoscopy was more difficult to perform through the 8th intercostal space compared with the 9th and 10th intercostal spaces. The 8th rib is the last sternal rib, and distracting this rib to advance the endoscope cranially created some discomfort as evident by slight movements back and forth is these cows. Because of the short distance from the 10th intercostal space to the diaphragm there was a restriction of the range of motion of the endoscope.

One cow had mild coughing during CO2 insufflation, which stopped when negative pressure was reestablished. Another cow had a mild tremor in the pelvic limbs during the initial 9th intercostal space thoracoscopic examination but not in subsequent examinations. One cow had insufficient lung collapse (right hemithorax after passive lung collapse, 9th intercostal space) because of local adhesions between the costal and the pulmonary pleura and pulmonary parenchymal injury with mild bleeding, but no air leakage was observed after trocar insertion. During the next examination after CO2 insufflation the trocar was introduced very carefully to avoid another laceration of the lung parenchyma. The adhesions were not disrupted by insufflation.

All cows recovered uneventfully and no postoperative complications or behavioral changes, signs of pain or discomfort were observed. There was no emphysema and apart from minor swelling there were no other signs of local inflammation at the portal sites. No abnormalities were detected during postoperative physical examinations and no residual pneumothorax was evident on radiographs 24 hours after the last thoracoscopic examination.

Examination After Passive Lung Collapse

The dorsolateral area of the collapsed cranial and caudal lobes of the lung, and the typical lobular pattern was readily visible during left and right thoracoscopy (Figs 1 and 2). Bilaterally, the collapsed lung was attached dorsally to the mediastinum by the pulmonary ligament (2). The proximal part of the 3rd–12th ribs and the internal intercostal muscles were visible on the left and right sides. The artery and vein of the neurovascular bundle on the caudal aspect of each rib, but not the intercostal nerve, was also visible.

image

Figure 1. Left pleural space, 9th intercostal space with the telescope advanced dorsocranially. LL, left lung; A: thoracic aorta; white arrow (solid line): left costocervical vein; white arrow (dotted line): left costocervical trunk; IM, intercostal muscles; LC, longus colli muscle; AZ, left azygos vein; 1, 2, 3: dorsal intercostal veins of ribs 3, 4, 5; DV, dorsal vagus nerve; III, IV, 3th, 4th rib; ML, middle tracheobronchial lymph node; red arrows: aortic vasa vasorum; E, esophagus.

Download figure to PowerPoint

image

Figure 2. Right pleural space, 9th intercostal space with telescope advanced dorsocranially RL, right lung; IM, intercostal muscles; LC, longus colli muscle; AZ, right azygos vein; 1, 2, 3: dorsal intercostal veins of ribs 2, 3, 4; II, III, IV, 2nd, 3th, 4th rib; arrows: pulmonary ligament.

Download figure to PowerPoint

Table 1. Anatomic Structures Visible During Left and Right Standing Thoracoscopy
Anatomic StructureLeft HemithoraxRight Hemithorax
  • *

    Under CO2 insufflation.

Pleural surfaces of ribs 3rd through 12thYesYes
Intercostal musclesYesYes
Longus colli muscleYesYes
Sympathetic nerve fibers (stellate ganglion)YesNot consistently seen
Left costocervical trunkYesNo
Right costocervical trunkNoNot consistently seen
Left costocervical veinYesNo
Right costocervical veinNoNot consistently seen
Brachiocephalic trunkYesNo
Dorsal and lateral lung surfacesYesYes
Middle and accessory lobe (parts)NoYes
Diaphragmatic surface of the caudal lung lobe (parts)YesYes
Pulmonary ligamentYesYes
Bronchoesophageal artery and veinNot consistently seenNot consistently seen
Oesophagus (thoracal part)YesYes
Aorta (thoracal part)YesYes
Azygos veinYesYes
Dorsal intercostal veinsYesYes
Dorsal intercostal arteriesYesYes
Pulmonary veinsNoNo
Thoracic ductNoNo
Thoracic aortic and intercostal lymph nodes (thoracic dorsal lymph center)YesYes
Caudal mediastinal lymph nodesYesYes
Sympathetic trunk (thoracic part)YesYes
Vagus nerve (dorsal branche)YesYes
Vagus nerve (ventral branche)NoNo
Recurrent laryngeal nerveNoNo
Phrenical nerveNoNo
Phrenical vein and branchesYesYes
Psoas major muscleYesYes
Diaphragm (parts)YesYes
Oesophagal hiatusNoNo
Aortic hiatusYesYes
Lumbocostal archYesYes
Heart base (pericardial adipose tissue)Yes (3/15)*Yes (1/15)*

On the left side, the thoracic aorta was centrally located and visible as were the aortic vasa vasorum. The left azygos vein appeared as a dark blue vessel crossing the thoracic aorta and coursing dorsally and parallel to it and receiving the dorsal intercostal veins caudal to each rib (1). The right azygos vein and its dorsal intercostal veins, varying in length, were visible in all cows. Four cows had this vessel extending to the 3rd rib, 4 cows to the 4th rib, and 7 to the 12th rib (2). By advancing the telescope more cranially within the left hemithorax, the brachiocephalic trunk could be seen originating from the aorta and continuing cranially. The left costocervical trunk, originating from the brachiocephalic trunk, was visible coursing dorsally similar to the left costocervical vein (1). Craniodorsally, the thoracic part of the longus colli muscle and sympathetic nerve bundles of the stellate ganglion were visible.

Because of superimposition of the lung, it was not always easy to identify the cranially located structures in the right hemithorax. On the left side, the tracheobronchial lymph node was observed between the thoracic aorta and the dorsal part of the left lung (1). Ventral to the aorta, the esophagus was visible and easily recognizable by its red color (1). During swallowing, the esophagus could be clearly identified, especially in the left hemithorax. Dorsal to the esophagus, the dorsal branch of the vagus nerve was visible as a thin white band; however, the ventral branch was not visible. Ventral to the heads of the ribs, the thoracic part of the sympathetic trunk was identified running horizontally as a thin white band. Lymph nodes were visible dorsal and ventral to the aorta (3). Moving the endoscope dorsocaudally, the thoracic aorta could be followed to the aortic hiatus at the lumbar part of the diaphragm. In the caudal mediastinum, the large caudal mediastinal lymph node was located ventral to the thoracic aorta. Consistently, the psoas major muscle was visible in the dorsocaudal pleural space (4). When directing the endoscope more caudally, the costal attachment of the diaphragm to the rib cage could be seen.

image

Figure 3. Left pleural space, 10th intercostal space with telescope advanced dorsomedially A: thoracic aorta; 1, 2, 3: dorsal intercostal veins of ribs 8, 9, 10; IM, intercostal muscles; AZ, left azygos vein; VIII, IX, X, head of ribs 8, 9, 10; ST, thoracic portion of sympathetic trunk; black arrow: lymph node (belongs to the dorsal thoracic lymph center); white arrows: dorsal intercostal arteries.

Download figure to PowerPoint

image

Figure 4. Left pleural space, 10th intercostal space with telescope advanced dorsocaudally ST, thoracic part of sympathetic trunk; A, thoracic aorta; CM, caudal mediastinal lymph node; black arrow: aortic hiatus; LP, lumbar part of diaphragm; TC, tendinous center of diaphragm; PM, psoas major muscle; XII, 12th rib; white arrow: costal attachment of diaphragm to the rib; black arrows (dotted line): branches of phrenic vein.

Download figure to PowerPoint

Moving the endoscope ventrally or slightly ventrocranially, the caudal edge of the collapsed lung was observed gliding over the pleural surface of the diaphragm (5). In the 10th intercostal space, advancing the endoscope between the caudal lung lobe and the diaphragm made portions of the diaphragmatic surface of the caudal lung lobe visible. In 1 cow, during examination of the right hemithorax, access to the middle and accessory lung lobes as well as the pericardial adipose tissue was possible, but only when the endoscope was directed forward between the diaphragm and the caudal lung lobe. Depending on the degree of lung collapse, the diaphragmatic tendinous center as well as the branches of the phrenic vein were visible (Figs 4 and 5). Thoracoscopic examination was concluded by advancing the endoscope in the ventrocranial direction. The pericardium was not visible in any examination with passive lung collapse.

image

Figure 5. Right pleural space, 10th intercostal space with telescope advanced slightly ventrocranially. RL, right lung; TC, tendinous center of diaphragm; IX, 9th rib; black arrow: phrenic vein; white arrows: acute margin of right caudal lung lobe; IM, intercostal muscles.

Download figure to PowerPoint

Examination After CO2 Insufflation

Insufflation to 5 mmHg subjectively increased lung collapse and slightly improved observation of the ribs and diaphragm in each hemithorax. CO2 insufflation did not improve visibility in the right cranial pleural space; however, in 3 cows, the pericardium became visible in the left hemithorax (2 from the 8th intercostal space and 1 from the 9th intercostal space; 6). Inspection of the pericardium was possible in 1 cow from the 9th right intercostal space. CO2 insufflation did not significantly improve visibility of the pericardium from either side (P > .05).

image

Figure 6. Left pleural space, CO2 insufflation, 8th intercostal space with telescope advanced ventrocranially. CRL, caudal part of the cranial lobe of the left lung; CAL, caudal lobe of the left lung; IM: intercostal muscles; V, 5th rib; H, auricle surface of heart; white arrow (dotted line): acute margin of the left cranial lung lobe; white arrow (solid line): caudal interlobular fissure.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIAL AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. DISCLOSURE
  8. REFERENCES

Thoracoscopic examination was successfully performed by using a 30°, 57-cm long, and 10-mm diameter rigid endoscope with video assistance in standing healthy adult Holstein–Friesian cattle. Restraint in stocks and use of local infiltration anesthesia at portal sites was sufficient for thoracoscopic examination. Some discomfort evident by the cow shifting back and forth was only observed when the endoscope was advanced cranially in the 8th intercostal space. This ceased when the endoscope was moved in a ventral, dorsal or caudal direction. Although not needed in these cows, supplemental analgesia and sedation could be provided by use of intravenous (IV) xylazine in combination with an intercostal nerve block. However, Peroni et al.[4] reported that despite use of regional and local anesthesia in combination with IV detomidine in horses, signs of pain occurred when a thoracoscopic approach in the 8th intercostal space was used. Because of the risks associated with general anesthesia and the development of tympany because of recumbency, surgical procedures should preferentially be done in standing cattle using local anesthesia.[3, 5, 6] Thoracoscopic pericardiotomy in a cow was performed solely using local anesthesia.[3]

Before inserting the trocar/cannula system into the pleural space we used a teat cannula to create pneumothorax with the intention of avoiding visceral damage.[3] In our opinion, this is a safe method. In 1 cow, trocar insertion resulted in a small pulmonary parenchymal injury with slight hemorrhage, because of adhesions between the costal and the pulmonary pleura; however, thoracic exploration was continued uneventfully. Inadequate lung collapse leading to a pulmonary parenchymal injury has been reported during equine thoracoscopy[4] and in small animals.[7]

From our experience, adequate examination of the intrapleural structures can be best achieved using the 9th intercostal space with the portal located at the level of the ventral margin of the tuber coxa. Although the 8th intercostal space might be preferable for examination of the cranial pleural space, it is less desirable for general thoracoscopic examination because the reduced flexibility of the sternal ribs contributes to discomfort during examination, a phenomenon also reported in equine thoracoscopy.[4, 8] Successful use of the 8th intercostal space was reported for pericardiotomy in a cow with lymphoma.[3]

Minimal side effects were observed in these cows. One cow had mild coughing during insufflation; however, this ceased when negative pressure was reestablished. Several reports describe coughing during thoracoscopic interventions in horses and people.[8-13] We recommend use of cannulas with a side stopcock, to which a suction tube can be attached to facilitate rapid control of intrapleural pressure and relief of pneumothorax if needed. One cow had mild pelvic limb tremor that only occurred during the initial thoracoscopic examination and no other discernible signs of discomfort or agitation were observed.

Standing thoracoscopy provided good viewing of the dorsal and lateral structures of the thorax. Collapsed lung, aorta, esophagus, diaphragm, and azygos vein were readily viewed; however, the pericardial area was not consistently visible. The thoracic duct, pulmonary veins, and esophageal hiatus of cattle were not visible, contrary to findings in horses.[4, 8] One reason for this may be that the increased connective tissue contained in the bovine mediastinum results in less mediastinal transparency than in horses.[14] The pulmonary veins and esophageal hiatus were obscured by the collapsed lung. We do not know if a higher insufflation pressure would have improved visibility of these structures or if withholding food preoperatively would improve overall intrathoracic observation.

CO2 insufflation of the pleural space performed under general anesthesia during thoracoscopy in horses and dogs does improve intrathoracic viewing.[15-17] In standing cattle, CO2 insufflation to 5 mmHg did not improve viewing of intrathoracic structures and particularly not in the right cranial thoracic space. One possible explanation might be that the right bovine lung is larger than the left and, therefore, occupies a wide part of the right pleural space despite being collapsed.[14] Insufflation to 5 mmHg made endoscopic inspection of the heart and pericardial tissue possible in four cows. In 1cow, the heart was visible on the right side without insufflation, but only when the endoscope was directed between the diaphragm and caudal lung lobe.

One limitation of our study is the absence of preoperative thoracic radiographs to rule out the presence of some pathologic findings, like masses, thoracic effusions, inflammatory pulmonary diseases amongst others that could have influenced our results, although other abnormalities like adhesions would likely not be detected. Another limitation was the order of examinations, which was always identical. In our experience with horses, beginning bilateral thoracoscopy on the left side reduces the risk of development of bilateral pneumothorax.[8] For this reason, and because we did not know if there might be similar findings in cattle, we decided to begin the procedure on the left side. We did not administer antibiotics despite using the same portal within 24 hours, because these were classified as clean procedures. No postoperative signs of inflammation or infection were observed.

Summarily, we found that bovine thoracoscopy can be safely performed on healthy standing cattle without major complications. Because cattle have an intact mediastinum,[14] the risk for respiratory problems during thoracoscopy (e.g., bilateral pneumothorax) is considered to be low. Thoracoscopic examination is minimally invasive method and should be considered as a complementary technique to visually explore the intrapleural space, obtain fluid and tissue samples, and to detect and diagnose pulmonary diseases.

DISCLOSURE

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIAL AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. DISCLOSURE
  8. REFERENCES

The authors report no financial or other conflicts related to this study.

REFERENCES

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIAL AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. DISCLOSURE
  8. REFERENCES
  • 1
    Mackey VS, Wheat JD: Endoscopic examination of the equine thorax. Equine Vet J 1985; 17:140142
  • 2
    Steiner A: Einsatz bildgebender Verfahren beim Rind. Tierarztl Prax Ausg G Grosstiere Nutztiere 2011; 39:275
  • 3
    Van Biervliet J, Kraus M, Woodie B, et al: Thoracoscopic pericardiotomy as a palliative treatment in a cow with pericardial lymphoma. J Vet Cardiol 2006; 8:6973
  • 4
    Peroni JF, Horner NT, Robinson NE, et al: Equine thoracoscopy: normal anatomy and surgical technique. Equine Vet J 2001; 33:231237
  • 5
    Hendrickson DA: Rib resection and pericardiotomy. in Hendrickson DA (ed): Techniques in Large Animal Surgery (ed 3). Ames, IA, Blackwell Publishing, 2007, pp 282286
  • 6
    Ducharme NG, Fubini SL, Rebhun WC, et al: Thoracotomy in adult dairy cattle 14 cases (1979–1991). J Am Vet Med Assoc 1992; 200:8690
  • 7
    Walton RS: Video-assisted thoracoscopy. Vet Clin North Am Small Anim Pract 2001; 31:729759
  • 8
    Scharner D: Thorakoskopische Untersuchungen am stehenden sedierten Pferd—Anatomie und Technik. Pferdeheilkunde 2012; 28:4652
  • 9
    Raphel CF, Gunson DE: Percutaneous lung biopsy in the horse. Cornell Vet 1981; 71:439448
  • 10
    Venner M, Schmidbauer S, Drommer W, et al: Percutaneous lung biopsy in the horse: comparison of two instruments and repeated biopsy in horses with induced acute interstitial pneumopathy. J Vet Intern Med 2006; 20:968973
  • 11
    Fair JJ: Anesthesia for thoracoscopy: an overview. AANA J 1994; 62:133138
  • 12
    Unzueta MC, Sabaté S, Casas JI, et al: Anesthesia for thoracoscopy. Rev Esp Anestesiol Reanim 1996; 43:6769
  • 13
    Dave N, Fernandes S: Anaesthetic implications of paediatric thoracoscopy. J Minim Access Surg 2005; 1:814
  • 14
    Budras K-D, Wünsche A: Brusthöhle. in Budras K-D, Wünsche A (eds): Atlas der Anatomie des Rindes. (ed 2). Hannover, Germany, Schlüterche Verlagsgesellschaft, 2002, pp 6265
  • 15
    Vachon AM, Fischer AT: Thoracoscopy in the horse: diagnostic and therapeutic indications in 28 cases. Equine Vet J 1998; 30:467475
  • 16
    Quandt JE: Anesthetic considerations for laser, laparoscopy, and thoracoscopy procedures. Clin Tech Small Anim Pract 1999; 14:5055
  • 17
    Walsh PJ, Remedios AM, Ferguson JF, et al: Thoracoscopic versus open partial pericardectomy in dogs: comparison of postoperative pain and morbidity. Vet Surg 1999; 28:472479