Technique and Peri-Operative Management of Left Pneumonectomy in Neonatal Piglets

Abstract

Background

Although commonly performed in adult swine, unilateral pneumonectomy in piglets requires significant modifications in the surgical approach and peri-operative care due to their smaller size and limited physiologic reserve.

Methods

Nineteen neonatal piglets underwent a left pneumonectomy. They were allowed 5–7 days of pre-operative acclimation and nutritional optimization. Pre-operative weight gain and laboratory values were obtained before the time of surgery. A “ventro-cranial” approach is adopted where components of the pulmonary hilum were sequentially identified and ligated, starting from the most ventral and cranial structure, the superior pulmonary vein. The principle of gentle ventilation was followed throughout the entire operation.

Results

The median age of the piglets at the time of surgery was 12 (10–12) days. The median pre-operative weight gain and albumin level were 20 (16–16)% and 2.3 (2.1–2.4) g/dL, respectively. The operation required a median of 59 (50–70) minutes to complete. Five of the first 9 piglets died from complications, two from poor pre-operative nutritional optimization (both with less than 10% weight gain and 2 g/dL for albumin), one from an intubation complication, one from intra-operative bleeding, and one in the post-operative period from a ruptured bulla. No mortality occurred for the next 10 cases.

Conclusions

Successful outcomes for unilateral pneumonectomy in piglets require special attention to pre-operative nutritional optimization, gentle ventilation, and meticulous surgical dissection. Pre-operative weight gain and albumin levels should be used to identify appropriate surgical candidates. The “ventro-cranial” approach allows for a technically straightforward completion of the procedure.

Introduction

Unilateral pneumonectomy is a commonly performed procedure in swine, especially in the field of transplantation research^1–7. Given the presence of an additional tracheal bronchus on the right side, left pneumonectomy has been preferred over right pneumonectomy due to simpler anatomy⁸. Most studies that utilize unilateral pneumonectomy do not describe the procedure and care in detail. Bufalari et. al. published the first report that comprehensively described the technique in adult swine⁹. Pneumonectomy performed in piglets requires significant modifications to the surgical approach due to their smaller size and limited physiologic reserve. Pre-operative optimization of neonatal piglets is more demanding with respect to nutrition and environmental adaptation. In this report, we aim to provide a thorough description for the technique of left pneumonectomy in neonatal piglets and guidelines for perioperative care.

Material and Methods

Pre-Operative Care

Nineteen female Yorkshire piglets (Parsons Pig Farm, Hadley, MA) underwent a left pneumonectomy. All procedures were carried out according to the National Institute of Health’ Guide for the Care and Use of Laboratory Animals and approved by the Institutional Animal Care and Use Committee at Boston Children’s Hospital. Three to six day old piglets arrive at our facility 5–7 days before surgery for acclimation and pre-operative optimization. The animals are housed together in groups of two or three with access to enrichment devices and water ad lib. The piglets are bottle-fed with warmed Grade A Ultra 24 Milk Replacer (Milk Products, Chilton, WI) four times per day, ensuring that no more than 12 hours elapse between feeds. Piglets often develop diarrhea during the transition from breast milk to formula. In order to avoid diarrhea, the formula concentration and volume to be administered are carefully and gradually adjusted. The initial feeding regimen consists of a maximum of 5 ounces per feeding of formula prepared as 1 cup of formula powder in 6 cups of water. The concentration is increased over a period of 6 days to 1.5 cups of formula in 6 cups of water and a maximum of 5–8 ounces per feeding (Table 1). In cases in which diarrhea develops, formula concentration can be reduced and oral aluminum/magnesium hydroxide (1–2 cc) is administered prior to each feeding. For piglets that develop severe diarrhea, oral metronidazole is added to the treatment regimen at a dose of 62.5 mg orally twice daily. Piglets are made nothing by mouth overnight in preparation for the surgical procedure on the following morning. Dry weight, complete blood count, and electrolytes are measured immediately before surgery.

Table 1.

Pre-operative and post-operative feeding regimen

Day	Concentration (cups of formula in 6 cups of water)	Maximal Volume (ounces)
Pre-operative
1–2	1	5
2–4	1.5	5
4–6	1.5	5–8
Post-operative
1–7	2	8–10
7–14	2	10–14

Anesthesia and Intubation

Pre-operative induction is achieved with an intramuscular injection of atropine (0.04 mg/kg), tiletamine-zolazepam (4.4 mg/kg), and xylazine (2.2 mg/kg). Anesthesia is maintained with isoflurane (1–4%) via facemask. Tracheal intubation is performed with a 2–4 mm cuffed endotracheal tube and a #1 Miller blade fiber-optic laryngoscope (Welch Allyn, Skaneateles Falls, NY). An intravenous catheter is placed in the marginal ear vein for fluid maintenance during the procedure. Hair over the left chest is removed with standard electric clippers. Vital signs including oxygen saturation percentage (SpO₂), heart rate, respiratory rate, and temperature are collected at baseline in preparation for the procedure.

The piglet is then transported into an operating room, placed on a heated surgical table, and mechanically ventilated and maintained on inhaled isoflurane (1–4%) with delivery via O₂/room air combination. Ventilator settings are adjusted to maintain peak airway pressures between 10–20 cmH₂O and end-tidal CO₂ between 30–40 mmHg. Normal saline is infused throughout the procedure at a rate of 5 ml/kg/hour to account for insensible losses. Electrocardiogram (ECG) and SpO₂ probes are placed on the tail for continuous monitoring. Heart and respiratory rate, temperature, SpO₂, peak airway pressure, end-tidal CO₂, tidal volume, and the ECG waveform are continuously monitored throughout the procedure and recorded every 15 minutes.

Left pneumonectomy

Figure 1 demonstrates the anatomy of the left pulmonary hilum with the right lung in the background. The piglet is placed in the right lateral decubitus position. A roll is inserted under the chest to widen the intercostal spaces. The forelimbs and hindlimbs are then anchored in place either by tape or ropes at a 45-degree angle to the body. The planned incision is located at the fourth intercostal space. With the lower edge of the thirteenth rib and twelfth intercostal space as starting landmarks, the fourth intercostal space is located by palpating cranially along the rib cage (Figure 2A). Typically, the fourth intercostal space lies approximately 6–7 cm caudal to the tip of the scapula. The entire rib cage to the level of the scapula is prepped with povidone-iodine and 70% ethanol. The primary surgeon stands on the ventral side of the animal and the assistant on the dorsal side. The required instruments are detailed in Table 2.

Figure 1. Sagittal view of the left pulmonary hilum as seen from the left side and with the right lung in the background.

Figure 2. Procedure of left pneumonectomy.

A: Incision marking (I), tip of the scapula (S), and inferior portion of the thirteenth rib (R)

B: Isolation of the superior pulmonary vein

C: Isolation of the pulmonary artery

D: Isolation of the left main bronchus

E: Isolation of the inferior pulmonary vein

F: Empty left thoracic cavity showing the (1) aorta, (2) azygous vein, (3) vagus nerve, (4) phrenic nerve, and (5) bronchial stump

Table 2.

Required surgical instruments

Instrument	Quantity
Scalpel (#15 blade)	1
Bovie electrocautery	1
Adson forceps	2
DeBakey forceps	2
Curved hemostat	1
Finocchietto retractor	1
Right-angled clamp	1
Right-angled dissectors	1
Metzenbaum scissors	1
Straight Mayo scissors	1
Needle driver	1
Skin stapler	1
Yankauer suction tip	1
Sutures
Silk (0, 2-0, and 3-0)
PDS® (2-0 and 3-0)

After skin incision is made, subcutaneous and muscular layers are sequentially divided with electrocautery. An intercostal nerve block is performed by injecting 2.5 mg/mL bupivacaine to the inferior surface of the fourth and fifth ribs at a total dose of 1 mg/kg. Electrocautery dissection is continued until the parietal pleura is identified. Blunt dissection with a curved hemostat is used to gain access into the thoracic cavity. The anesthesiologist is notified at this point to lower the tidal volume to avoid inadvertent injuries to the lung underneath.

A Finocchietto retractor is then inserted between the fourth and fifth ribs to provide full exposure of the thoracic cavity. A correctly located thoracotomy incision allows for direct visualization of the major fissure of the left lung. The superior lobe is then gently isolated and reflected caudally with a right-angle clamp. This maneuver allows access to the pulmonary hilum and a view of its ventro-cranial aspect. The assistant maintains gentle retraction on the lung while the primary surgeon isolates all hilar structures with blunt dissection and a right-angled dissector. Dissection of the hilum begins with identification of the superior pulmonary vein (PV), which is the most ventral and cranial structure in the hilum (Figure 2B). The PV lies just caudal to the azygous vein as it empties into the superior vena cava and often consists of two branches that require separate ligations. Silk ties (Ethicon, Somerville, NJ) are used for vessel ligation. Division of the superior PV allows access to the pulmonary artery (PA), which is dorsal and caudal to the PV (Figure 2C). The vagus nerve courses along the dorsal aspect of the pulmonary hilum and tends to be adherent to the PA. It should be identified and dissected free of the PA to avoid inadvertent injury. Similarly, the identification and careful manipulation of the phrenic nerve, which courses along the ventral aspect of the pulmonary hilum, will avoid its inadvertent injury that leads to diaphragmatic paralysis. Once the PA is dissected off of the left main bronchus, it is ligated with silk ties and divided. The left main bronchus is caudal to the PA and easily identified by its larger size and cartilaginous consistency (Figure 2D). The inferior PV is caudal to the left main bronchus and often hidden from view due to the size of the bronchus. Therefore dissection is kept close to the bronchus to avoid injury to the inferior PV. The bronchus is ligated with a silk tie. At the time of bronchus ligation, the anesthesia team is asked to lower the tidal volume and ensure that peak airway pressures remain between 10–20 mmHg. The bronchus is then clamped just distal to the previously placed silk tie and sharply divided with a scalpel over the clamp. A major air leak or bronchial stump blowout is prevented by placing an additional silk suture ligature under the clamp. The most caudal hilar structure, the inferior PV, will come into view following ligation and division of the bronchus (Figure 2E). The inferior PV is usually larger than the superior PV and runs in continuation with the inferior pulmonary ligament. The inferior pulmonary ligament is a thin avascular membrane that tethers the medial surface of the lower lobe of the lung to the mediastinum. The inferior PV is safely separated from the inferior pulmonary ligament by dissecting close to the lung and avoiding injury to the mediastinal pleura. The inferior PV is then ligated with silk ties and divided.

The assistant surgeon maintains gentle cranial traction on the lung while the primary surgeon divides the inferior pulmonary ligament with Metzenbaum scissors, remaining close to the lung and avoiding injuries to the mediastinal pleura. After the specimen is removed, the empty left thoracic cavity (Figure 2F) is filled with warm saline to check for both air leak and bleeding. Large breaths are avoided during this step to avoid barotrauma to the contralateral lung. Instead, the animal is ventilated normally as this is sufficient to detect any physiologically significant air leak. If the mediastinal pleura is injured during division of the inferior pulmonary ligament, air evacuation is achieved with a red-rubber catheter inserted into the contralateral thoracic cavity through the communication. The catheter’s end is exteriorized through a separate small skin incision and a “U” stitch with PDS^® (Ethicon, Somerville, NJ) suture is then fashioned around the tube in preparation for its removal.

The thoracotomy incision is closed with multiple interrupted PDS^® sutures after removal of saline. Care is taken to avoid the neurovascular bundles along the caudal surface of the rib. The knots are tied once the chest roll is removed. The muscular layer is closed with running PDS^® and the skin is closed with staples (Covidien, Dublin, Ireland) (Figure 2H). In case of violation of the mediastinal pleura, 20–40 mL of air is evacuated through the previously placed red-rubber catheter. The catheter is placed under water seal and removed while the anesthesia team maintains a gentle breath hold. Care is taken not to remove an excessive volume of air as this could create a vacuum effect and result in displacement of the mediastinal structures. The “U” stitch is tied down to seal the track and surgical wounds are covered with Ioban™ drapes (3M, Saint Paul, MN).

Post-Operative Recovery

Extubation occurs once the piglet achieves spontaneous respiration, and sufficient end-tidal CO₂ (30–35 mmHg), SpO₂ (>60%), and pCO₂ (<60 mmHg). The animal is transported to an incubator where ambient temperature is maintained between 30–32 °C to avoid hypothermia. The piglet’s recovery is monitored at all times. Piglets may become agitated as they emerge from anesthesia. During the recovery, common signs of distress such as shivering (hypothermia), increased agitation (pain or hypoxemia), pallor (hypoxemia), and prolonged somnolence (hypoglycemia) are promptly recognized and treated. Hypothermia is addressed by increasing the ambient temperature. Hypoxemia is treated with supplemental O₂ delivered via a facemask. Hypoglycemia is usually managed by 3 mL of 10% dextrose in water by mouth. Analgesia is administered in the form of a fentanyl patch at 4–5 µg/kg placed immediately after the procedure. Full recovery is achieved when the animal regains balance and is independently stable in the upright position. At this point bottle-feeding may resume to prevent hypoglycemia and nutritional deficiencies.

Post-Operative Care

The animal is transferred from the incubator to its pen on post-operative day (POD) 1 provided it continues to show no sign of distress. Post-operative prophylactic antibiotics consist of oral amoxicillin at 20 mg/kg twice daily for 7 days. If the animal continues to have diarrhea, amoxicillin is substituted with oral metronidazole at 62.5 mg twice a day. The analgesia regimen consists of a fentanyl patch for maintenance and intramuscular flunixin meglumine at a dose of 2 mg/kg once per day as needed for breakthrough pain. The animal continues to be bottle-fed four times per day as described above. Piglets should tolerate as much as 8–10 ounces of formula per feed by POD 7 and 10–14 ounces per feed by POD 14 (Table 1). After POD 14, the animal is introduced to creep feeds (Mini Pig Starter Diet, Scott’s Distributing, Hudson, NH) and the frequency of formula feeds can be decreased to 2–3 times per day. Post-operative weight is measured periodically every 2–3 days and a necropsy is performed on all animals at the time of death or euthanasia. Animals in this study are euthanized between 1 to 3 weeks after the time of surgery with sodium pentobarbital at a dose of 110 mg/kg.

Data Analysis

Descriptive variables are expressed as median with interquartile range (IQR). Data analysis was performed with Prism (Graphpad Software, La Jolla, CA).

Results

The piglets were 6 (3–6) days old at the time of arrival (Table 3). Over 5 to 7 days of pre-operative optimization and acclimation, the piglets achieved a net gain of 20 (16–26) % of their arrival body weight. Median age at time of surgery was 12 (10–12) days old. Pre-operative laboratory values are listed in Table 2. The median albumin level was 2.3 (2.1–2.4) g/dL.

Table 3.

Pre-operative data, operative duration, and post-operative data for piglets undergoing left pneumonectomy

	Median (IQR) N = 17*
Age at arrival (days)	6 (3 – 6)
Age at operation (days)	12 (10 – 12)
Weight at operation (kg)	2.8 (2.4 – 2.9)
Percent weight gain (%)	20 (16 – 26)
Pre-operative laboratory values
White blood cell count (×103/µL)	6.4 (5.3 – 6.9)
Hemoglobin (g/dL)	9.7 (8.1 – 10.4)
Hematocrit (%)	34.0 (29.3 – 35.8)
Platelet (×103/µL)	425 (364 – 460)
Albumin (g/dL)	2.3 (2.1 – 2.4)
Sodium (mEq/L)	137 (136 – 138)
Potassium (mEq/L)	4.2 (4.0 – 4.6)
Blood urea nitrogen (mg/dL)	5.0 (3.0 – 9.5)
Creatinine (mg/dL)	0.7 (0.6 – 0.9)
Operative duration (minutes)	59 (50 – 70)
Post-operative venous blood gas values
pH	7.43 (7.41 – 7.46)
pCO2 (mmHg)	59 (55 – 69)
pO2 (mmHg)	51 (44 – 56)
Percent oxygen saturation	65 (61 – 84)
Bicarbonate (mEq/L)	40.8 (38.2 – 43.9)
Post-operative percent weight gain (%)§
Post-operative day 7	11 (4 – 22)
Post-operative day 14	95 (85 – 102)
Post-operative day 21	115 (106 – 149)

^*We excluded the first 2 piglets that died from poor pre-operative optimization

^§Five piglets were weighed at POD 7, four at POD 14, and five at POD 21

Operative time required a median of 59 (50–70) minutes to complete (Table 3). The immediate post-operative median venous pCO2 was 59 (55–69) mmHg and pH was 7.43 (7.41–7.46). Post-operative weight gain was 11% on POD 7 and rapidly increased to 95% by POD 14 (Table 3).

Of the 19 piglets that underwent a left pneumonectomy, five died or required euthanasia soon after surgery (Table 4). Two animals died due to poor pre-operative nutritional optimization. They both had severe diarrhea that resulted in poor nutritional status, albumin < 2 g/dL and failure to gain weight. Two piglets died from intra-operative complications. One of the intraoperative deaths was attributed to significant blood loss from an inadvertent injury to the superior PV, while the other was the result of tracheal perforation during intubation that led to the animal’s death upon extubation. One animal died from an unexpected post-operative complication. The piglet was discovered in a state of severe respiratory distress with massive subcutaneous emphysema during the first post-operative check on the night of surgery. The animal was promptly re-anesthetized and underwent an emergent exploratory thoracotomy. An air leak test revealed an intact bronchial stump but the presence of a ruptured bulla on the diaphragmatic surface of the remaining right lung. The animal suffered cardiac arrest soon after induction despite maximal resuscitative efforts and died shortly afterwards. These 5 cases were encountered during our first 9 operations; no mortality occurred in the subsequent 10 cases. At necropsy, all piglets that survived until the end of the study period demonstrated various degrees of intrathoracic adhesions, with no evidence of significant bleeding or infection. Ligated structures in the pulmonary hilum appeared intact, with no evidence of bronchial stump blowout or hemorrhage. The contralateral lung showed evidence of compensatory growth.

Table 4.

Mortality cases of piglets undergoing left pneumonectomy

Case Number	Outcome	Cause of Death
1	Death on POD 0	Inadequate pre-operative nutritional optimization (5% weight gain, albumin 1 g/dL)
2	Euthanized on POD 1	Inadequate pre-operative nutritional optimization (10% weight loss, albumin1.1 g/dL)
3	Death on POD 0	Injury to superior PV causing excessive intra-operative blood loss (approximately 30 mL)
6	Death on extubation	Tracheal perforation during intubation
9	Death on POD 0	Ruptured bulla on diaphragmatic surface of the remaining right lung

Discussion

Although the model of unilateral pneumonectomy has been widely studied in adult swine, translating it to neonatal piglets requires technical modifications and detailed peri-operative care due to their limited size and physiologic reserve. Nutritional optimization is particularly important in determining post-operative survival. In our experience, piglets with a low albumin, <2 g/dL, and poor pre-operative weight gain, <10 %, demonstrate inadequate nutritional optimization and should not undergo an operative procedure.

Unilateral pneumonectomy places considerable stress on the animal’s cardiopulmonary system and hemodynamics, especially in the setting of severe blood loss^10,11. The combination of pneumonectomy and high blood loss can lead to severe right heart failure and rapid decompensation despite aggressive fluid resuscitation. For this reason, and given the fragile nature of the involved structures, a successful operation requires meticulous dissection of the pulmonary hilum. Care must be taken to avoid injuries to the vascular structures, the dorsally-located vagus nerve and ventrally-located phrenic nerve. Injuries to the phrenic and vagus nerves may lead to diaphragmatic and vocal cord paralysis, respectively, in addition to future abnormalities in airway development¹².

We preferred silk suture for ligation and suture ligature of the left main bronchus due to the ease of handling and short duration of post-operative follow-up. For piglets that are intended for longer survival, silk suture should be substituted with Vicryl^® (Ethicon, Somerville, NJ), PDS^®, or stapler. Some surgeons also argue that there is a theoretical risk of injuring the bronchial microvasculature and higher incidence of dehiscence because of its clamping. In our experience, all 19 piglets had intact bronchial stump at the time of death or euthanasia and clamping of the bronchus does not seem to increase the risk of stump blowout. Bronchial clamping in this procedure only serves to facilitate its division and placement of the suture ligature as an additional layer of protection from stump blowout. This step certainly can be omitted at the surgeon’s discretion. Due to its small size in neonatal piglets, it is impractical to place multiple interrupted sutures to close the bronchial stump. We therefore advocate for the placement of an additional suture ligature to prevent stump blowout especially due to lack of a hilar fat pad or substantial pleura for coverage.

Several studies advocate for the use of a heparin bolus before the clamping and division of hilar vessels^1,2,6. In our experience, this is not necessary as we avoid clamping or placement of additional running sutures upon the division of these vessels. The hilar vessels in the piglets are small and therefore running sutures are not necessary. A single 3-0 or 2-0 silk tie is sufficient for vascular control of the pulmonary veins and artery, respectively. Most studies also advocate for the dissection of all vascular structures before ligation and division. This is particularly difficult in a piglet model due to both space constraint and limited separation among these structures. For this reason, a “ventro-cranial” approach is taken where hilar structures are sequentially dissected and ligated in the order that they appear, starting from the superior PV. This approach appears to be safe and allow for easy control of hilar structures in a relatively short period of time. Additionally, the PA is identified and controlled early in the course of the operation, prior to division of the last PV. This has the likely advantage of minimizing organ engorgement and blood loss.

Communication between the surgical and the anesthesia teams is paramount to the success of this operation. The first point of critical communication occurs when the thoracic cavity is entered. At this time, lowering the tidal volume can avoid inadvertent injuries to the underlying lung during electrocautery division of the parietal pleura. The most critical checkpoint takes place upon ligation of the left main bronchus. The anesthesia team must be informed of this step both before and while the suture is tied. Ventilatory settings should be closely monitored during this stage of the operation to maintain peak airway pressures between 10–20 mmHg to avoid barotrauma to the contralateral lung, which is a potentially devastating complication. One death in our cohort was due to a ruptured bulla on the remaining right lung and was likely a result of unrecognized barotrauma. For this reason, it is our practice to adopt gentle ventilation throughout the operation and even avoidance of a large breath hold during the air leak test.

Unilateral pneumonectomy results in hemodynamic alterations in the remaining lung. There is a progressive increase in the pulmonary mean arterial pressure (MAP), PaCO₂, airway resistance, and peak airway pressure. Concurrently, there is a reduction in systemic MAP, cardiac index, and global ejection fraction^13,14. Interestingly, a right pneumonectomy results in more prominent changes in these parameters compared to left pneumonectomy¹⁴. The piglet’s position after surgery also has a significant impact on hemodynamic status. The supine position produces the most severe derangements in systemic MAP, cardiac index, and global ejection fraction while the lateral decubitus position with the remaining lung elevated produces the most favorable hemodynamic results^14,15.

This study was limited by its descriptive nature and a relatively small number of piglets. Other unexpected complications may arise when the procedure is expanded to a larger number of animals. Addressing those complications would further refine the technique and improve the mortality rate beyond what we have observed. Although we encountered five early mortalities in the first nine piglets, only one of these cases was the result of an intra-operative technical complication, i.e. injury to the superior PV. Two other piglets expired due to poor physiologic reserve while one died from an intubation complication and one succumbed to ruptured bullae after surgery.

Unilateral left pneumonectomy in neonatal piglets poses challenges due to the animals’ size and fragility. Survival is highly dependent on pre-operative nutritional optimization and a bloodless operation. Pre-operative weight gain and albumin levels should be used as indicators of adequate nutrition optimization and to screen for appropriate surgical candidates. Sequentially controlling the pulmonary hilum from a “ventro-cranial” approach provides the best exposure of all hilar structures and allows for a straightforward completion of the procedure.