A Comprehensive Protocol for Laparotomy in Swine to Facilitate Ultrasound-guided Injection into the Fetal Intraperitoneal Space

Abstract

Swine are a commonly used animal model for biomedical research. One research application of swine models is the in utero injection of human or pig cells into the fetal liver (FL) or intraperitoneal space. In utero injections can be accomplished through laparotomy procedures in pregnant swine. In this study, we aimed to establish comprehensive laparotomy protocols for ultrasound-guided injections into fetuses. Two pregnant gilts, with a total of 16 fetuses, underwent laparotomy at 41 and 42 d of gestation. During surgery, we attempted to inject half of the fetuses in the FL or intraperitoneal space with saline and titanium wire for radiographic imaging after birth. After the laparotomy and fetal injections, both gilts maintained pregnancy throughout gestation and initiated labor at full term. Of the 16 fetuses present at the time of laparotomy, 12 were liveborn, 2 were stillborn, and the remaining 2 were mummies. A total of 7 fetuses from the 2 litters were known to have been injected with a wire during the surgery. After farrowing, piglets were radiographed, and 6 piglets were identified to have wire within the abdominal space. Livers were dissected, and additional radiographs were obtained. It was determined that one piglet had wire within the liver, whereas the other 5 had wire within the intraperitoneal space. Overall, we describe in-depth laparotomy surgery protocols, ultrasound-guided injection of saline and titanium wire into the FL or intraperitoneal space, postoperative monitoring protocols, and information on radiographic detection of titanium wire after piglet birth. These protocols can be followed by other research groups intending to inject cells of interest into either the intraperitoneal space or FL of fetal piglets.

Swine are a popular animal model for biomedical research. Many research groups use pigs for cancer,^17,18 cardiovascular,²² and infectious disease¹¹ studies and other areas of research. An application of swine models that is gaining interest is the ability to study human immune cell function and de novo development of human cells—also known as ‘humanization’—in pigs. Previous humanization studies in swine have been performed through in utero injection of human hematopoietic stem cells (HSC) into the fetal intraperitoneal space.^4,10,13,14 Injections typically occur around 40 d of gestation, because this is the time period when hematopoiesis is occurring in the swine fetal liver, an organ to which injected cells migrate.¹⁹

Injections into the intraperitoneal space of fetal piglets can be accomplished through either percutaneous injection through the gilt's abdomen and into the uterus or laparotomy followed by injection through the uterine wall. One group injected human HSC by using ultrasound-guided percutaneous injection, given that laparotomy resulted in high rates of abortion in their study. In addition, fetuses from the cited study were injected with small titanium wires so that the injection site could be confirmed later by radiographic analysis.⁴ More recent human HSC injections in swine have combined laparotomy with ultrasound-guided fetal intraperitoneal injections.^10,13,14 In addition, laparotomy techniques are used for the injection of human liver cells into swine fetal livers.³ Laparotomy procedures allow for more precise aiming into the fetuses due to closer contact to the injection site.

Pigs with SCID are an emerging animal model.^{7,8,15,20,21,24,25} SCID pigs are a particularly important biomedical model to develop due to their ability to accept human xenografts, including pluripotent stem cells⁹ and cancer cells.² It is of interest to the biomedical field to generate humanized SCID pig models and to produce rescued SCID pigs that can be used for breeding purposes by using porcine HSC transplantation for the propagation of these animals. HSC transplantation can be achieved through ultrasound-guided injection into the fetal intraperitoneal space or fetal liver (FL) for the engraftment of human or porcine HSC. In addition, an interesting question to address is how HSC engraftment levels differ when cells are injected into the fetal intraperitoneal space compared with the FL. Titanium wire injection into these locations can be used for radiographic identification after piglet birth and as analysis of cell engraftment takes place.

Several research studies outline brief descriptions of swine laparotomy techniques;^3,4,10,13,14 however, no publications are dedicated solely to protocols for ultrasound-guided fetal injections through midline laparotomy of swine in mid-gestation. Here we detail safe and successful surgical procedures, postoperative monitoring protocols, injections into the fetal intraperitoneal space and FL, and radiographic results from farrowed piglets. We injected saline as a model for a cellular suspension and titanium wire for radiographic analysis. These protocols will be valuable to veterinary surgeons working with swine in biomedical settings, to help minimize losses, morbidity, and animal numbers. Through this work, we aim to provide protocols that can be referenced for consistency across different groups that perform laparotomies in swine for the injection of cells of interest into fetal piglets.

Materials and Methods

All protocols and methods were approved by Iowa State University's IACUC. All protocols and procedures performed were in accordance with The Guide for the Care and Use of Laboratory Animals.⁶

Animals.

Two commercial crossbred (Yorkshire×Duroc; age, approximately 1 y) gilts (referred to as 31-7 and 34-8 herein) were bred at a university-associated facility. At the time of surgery, gilt 31-7 weighed 167 kg and gilt 34-8 weighed 182 kg. Gilts were pregnancy-checked 28 d after breeding and confirmed to be negative by PCR analysis for porcine reproductive and respiratory syndrome virus and porcine circovirus type 2 before being transferred to Iowa State University Laboratory Animal Resources (ISU LAR) Facilities at gestational day 40. Gilt 31-7 underwent surgery at gestational day 41, whereas gilt 34-8 underwent surgery at gestational day 42.

Gilts were housed in clean conventional housing rooms at ISU LAR Facilities after surgery and throughout the rest of gestation. Specifically, the gilts were singly housed on epoxy-coated flooring with nose-to-nose contact with each other and mats covering part of the floor. Animals were given continual access to water, and limit-fed a custom swine gestation diet free of dried distiller's grains (Heartland Coop, Prairie City, IA), with chopped straw and timothy hay cubes added. Rooms were maintained at 68 to 72 °F (20.0 to 22.2 °C). At gestational day 109 gilts were moved to farrowing decks, with plastisol-coated metal-grate flooring; room temperature was increased to 73 to 75 °F (22.8 to 23.9 °C). After the gilts farrowed, the diet was changed from a gestation formulation to one for lactation. Heat lamps were placed on the sides of the farrowing crates to provide additional warmth for the piglets. All animals were checked every 12 h.

Laparotomy surgery protocol.

Starting 1 d before surgery, pregnant gilts undergoing laparotomy received 15 mg of altrenogest (Matrix, Merck Animal Health, Madison, NJ) in the morning feed (dosage independent of body weight) at approximately the same time each morning. Both gilts were maintained on altrenogest through gestational day 113. The feed was withheld from both animals 12 h prior to surgery. On the morning of surgery, altrenogest was given orally directly, instead of in the feed.

Prior to anesthesia, gilts were given glycopyrrolate (0.01 mg/kg IM; West-Ward Pharmaceuticals, Eatontown, NJ) and then anesthetized by using tiletamine–zolazepam (5 mg/kg IM; Zoetis, Parsippany, NJ) and xylazine (2 mg/kg IM; Akorn, Lake Forest, IL). After gilts were anesthetized, hair was clipped from the lower abdomen, and the surgical site was washed. Gilts were then intubated and began receiving isoflurane (2% to 5%) and oxygen (2.5 L/min). The gilts were then transferred to a surgical table and placed in dorsal recumbency. A 20-gauge intravenous catheter was placed in the ear vein. Lactated Ringer's solution (Hospira, Lake Forest, IL) was administered intravenously as a constant rate infusion for a total of 0.83 to 1.25 L/h. A presurgical scrub was performed with chlorohexidine (Phoenix, St. Joseph, MO). Sterile material drapes and Ioban (3M, Maplewood, MN) were used to cover the surgical field. Cefazolin (West-Ward Pharmaceuticals, Eatontown, NJ) was administered intravenously via the catheter starting within 10 min of the initial incision.

A ventral midline incision from the caudal-most nipple, extending cranially to the caudal aspect of the umbilicus, through the linea alba, and into the peritoneal cavity was made by sharp dissection. The left horn was exteriorized first; fetuses were visualized by ultrasonography, and a sterile gentian violet ink marker (Medline Industries, Dubuque, IA) was used to physically mark the outside of the uterus at each fetus (Figure 1 A); every other fetus in the horn was injected. The left horn was then replaced into the abdominal cavity. The same procedure was then performed on the right horn.

Figure 1.

Exteriorization of uterine horns for ultrasound-guided injection of wire and saline (A) During laparotomy, each horn was exteriorized to identify fetuses by ultrasonography. Fetuses were marked by using a water-soluble marker. (B) Ultrasonography was used to guide injection of wire and saline into fetuses. (C) An ultrasonogram of fetal liver. (D) A 22-guage needle was preloaded with a long copper-wire plunger. A 2-mm titanium wire was placed at the tip of the needle. The copper wire was then used as a plunger to push the titanium wire into fetal tissues.

Both uterine horns and the abdominal cavity were lavaged with a total of 500 mL of Lactated Ringer's solution. The abdominal cavity was closed at the linea alba in a simple continuous pattern (1 polydioxanone suture) with an additional overlying cruciate pattern (1 polydioxanone suture). The subcutaneous tissue was then closed with a simple continuous pattern (0 polyglycolic acid suture). The skin was closed in a buried subcuticular pattern (0 polyglycolic acid suture). The incision was covered with sterile gauze, which was held in place with an Ioban drape (3M). A list of all drugs, dosages, and equipment is found in Figure 2.

Figure 2.

Information regarding drugs and equipment used.

Ultrasound-guided injection of saline and wire into intraperitoneal space or fetal liver.

A ZONARE ultrasound system (Mindray, Malwah, NJ) with a 10-MHz intraoperative linear array transducer (model L14-5sp) was used to identify fetuses and guide injections. A titanium wire (diameter, 0.25 mm; length, 2 mm) was injected into the intraperitoneal space or FL by using a 22-gauge needle and a plunger made from multiple pieces of copper speaker wire. Copper wire plungers were preloaded into 22-gauge needles and autoclaved in sterile pouches with precut titanium wires.

As described earlier, the uterine horns were imaged by using ultrasonography to mark the number and location of fetuses present. Every other fetus in both horns was injected (Figure 1 B). Ultrasound guidance was used to identify either the FL (Figure 1 C) or intraperitoneal space. Once these areas were identified, titanium wire was loaded into the 22-gauge needle (Figure 1 D) by using fine-pointed curved forceps. The needle was then aimed for either FL or intraperitoneal space by keeping the needle parallel to the ultrasound transducer to allow for visualization. Once the needle was inserted in the appropriate location, the copper wire plunger was moved forward to push the titanium wire out and into the fetus. The copper wire was then removed from the needle while the needle was kept in the same place in the fetus. A 1-mL syringe containing 0.1 mL saline was inserted into the needle and the solution injected into the fetus (Figure 1 B).

Postoperative care of pregnant gilts.

After the Ioban bandage was placed over the incision, animals received sustained- release buprenorphine (0.18 mg/kg SC; ZooPharm, Windsor, CO); buprenorphine reaches therapeutic levels in the serum by 30 min after administration.^5,23 Then ceftiofur crystalline free acid (5 mg/kg IM; Excede, Zoetis, Parsippany, NJ) and meloxicam (0.3 mg/kg SC; Norbrook, Overland Park, KS) were administered. Gilts were then removed from the surgery table and placed in a recovery pen with warming blankets. Rectal temperatures were taken approximately every 30 min. A total of 2.5 L of Lactated Ringer's solution was given intravenously until the gilts began to wake.

Gilts were checked every 12 h for 14 d after surgery. Animals were monitored for evidence of adequate pain control according to attitude, appetite, vocalization, and locomotion. In addition, rectal temperatures, vaginal discharge, urination, and defecation were recorded. For the first 3 d after surgery, gilts received meloxicam (0.25 to 0.3 mg/kg PO, rounded to the nearest half 15-mg tablet) in the morning; they also received 15 mg of altrenogest (Merck Animal Health) dressed over their morning feed daily until gestational day 113. The Ioban bandage was removed 1 wk after surgery.

X-ray detection of injected wires.

Piglets from both litters were delivered at full term. Piglets nursed until 2 or 3 d of age and were then euthanized by using intravenous pentobarbital sodium (2 to 4 mL per piglet; Fatal-Plus, Vortech Pharmaceuticals, Dearborn, MI). Sows were euthanized by intravenously administering 60 mL of pentobarbital sodium. Piglets were then brought to the Iowa State University Small Animal Clinic, where lateral and dorsoventral radiographs were acquired from intact piglets, as well as dissected livers.

Results

Ultrasound-guided injection into the intraperitoneal space and liver of fetal pigs.

One goal we aimed to accomplish through injecting the fetal pigs was to determine the feasibility of specifically targeting the fetal intraperitoneal space or liver. To this end, fetuses were injected with a titanium wire at the site of saline injection such that X-ray analysis could be performed later to locate the site of injection. We aimed to inject the fetal intraperitoneal space in gilt 31-7 and the FL in gilt 34-8. We decided to only inject half of each litter, given that in the case that injections caused fetal death, the pregnancy would likely be maintained by surviving (noninjected) piglets. All fetuses are listed in Table 1.

Table 1.

Ultrasonographic overview of fetal injection locations during laparotomy of gilts 31-7 and 34-8

Uterine horn	Fetus	Injection location notes
Gilt 31-7
Right
	A	0.1 mL saline, intraperitoneal. Wire came out
	B	Not injected
	C	0.1 mL saline and wire, intraperitoneal
	D	Not injected
Left
	E	0.1 mL saline and wire, potentially in fetal liver
	F	Not injected
	G	0.1 mL saline and wire, intraperitoneal
	H	Not injected
	I	0.1 mL saline and wire, intraperitoneal
	J	Not injected
Gilt 34-8
Right
	K	Not injected
	L	0.1 mL saline and wire, in fetal liver
	M	Not injected
Left
	N	0.1 mL saline and wire, in fetal liver
	O	Not injected
	P	0.1 mL saline and wire, potentially intraperitoneal

Gilt 31-7 had a total of 10 fetuses, with 4 in the left horn and 6 in the right horn; we attempted to inject a total of 5 piglets with wire and saline. Saline and wire were injected into fetuses A, C, E, G, and I, but the wire of fetus A came out of the uterine wall immediately after the injection; thus fetus A is reported as saline-injected only. We did not attempt to inject another wire into fetus A. On the basis of ultrasonographic images obtained at the time of injection, we anticipated that we injected wire within the intraperitoneal space of 3 fetuses (C, G, and I) and within the FL of the remaining one (E). Gilt 34-8 had a total of 6 fetuses, with 3 in each horn; we injected a total of 3 fetuses with wire and saline. Of these, we anticipated that we injected wire within the FL of 2 fetuses (L and N) and the remaining one in the intraperitoneal space (P) (Table 1).

Postoperative monitoring.

Another important aspect of this surgery was to ensure that gilts did not succumb to infection or abort fetuses after the surgery was performed. Gilts were closely monitored during the 14 d following the surgery. Neither animal presented with a fever or vaginal discharge during this time period or for the rest of gestation. One gilt (34-8) showed moderate hindleg stiffness 2 d after surgery. Meloxicam was continued an additional 2 d; no pain was observed after the additional doses or at discontinuation of meloxicam. We noted no other signs of pain in the gilts: they presented with normal appetites, good locomotion, and bright attitudes. Bandages were replaced when they became loose or when incisional drainage was noted. Bandages were removed from both gilts 1 wk after surgery. Minimal redness and swelling were present around the incision sites. Altrenogest was administered each morning, to help prevent abortions. Both gilts were moved to farrowing crates at gestational day 109, before going into labor at full term.

Farrowing and radiographic identification of wire location post birth.

In addition, it was imperative to determine whether injected fetuses developed normally and that the gilts had gestated normally after laparotomy procedures. Piglets ID and birth weights are listed in Table 2. The numbering of the piglets in Table 2 does not correspond to that in Table 1 because we did not have a way to determine piglet location in the uterus as they were born.

Table 2.

Birth status of piglets from gilts 31-7 and 34-8

Piglet	Status	Birth weight (kg)	Wire location
Gilt 31-7
317-01	Live	1.3	Liver
317-02	Live	1.4	—
317-03	Live	1.3	—
317-04	Live	1.1	—
317-05	Live	1.3	—
317-06	Live	1.3	—
317-07	Live	1.0	Intraperitoneal
317-08	Live	1.2	Intraperitoneal
317-09	Live	1.4	—
317-10	Stillborn	1.1	—
Gilt 34-8
348-01	Live	1.0	—
348-02	Live	1.5	Intraperitoneal
348-03	Live	1.5	Intraperitoneal
348-04	Stillborn	1.2	—
348-05	Mummy	NA	—
348-06	Mummy	NA	Undetermined

The numbering of the piglets in Table 2 does not correspond to that in Table 1 because we did not have a way to determine piglet location in the uterus as they were born.

Gilt 31-7 (litter 317) farrowed 9 live and one fully developed stillborn piglet with no complications. After the first 3 piglets were manually assisted with delivery, gilt 34-8 (litter 348) required a Cesarean section (C-section) due to uterine inertia causing dystocia. The C-section took place approximately 5 h after the previous piglet was delivered (24 h after first milk streaming was observed). The C-section was performed by euthanizing the gilt and exposing the uterus to retrieve any other piglets that were present. One fully developed stillborn (348-04) and one mummy (348-06) were found. The 6th known piglet (348-05) was not found within the uterus; mummy was designated for this pig. We do not consider that the uterine inertia was due to the laparotomy procedures but rather to piglet malposition in the birth canal and resultant unproductive contractions. Piglets from the 348 litter were cross-fostered on sow 31-7 for the duration of the study.

Piglets were euthanized at 2 d (litter 317) or 3 d (litter 348) of age, at which point they were radiographed for the detection of wires. From the surgeries, we expected that 7 of the total 16 piglets (Table 1) would have wires in either the liver or intraperitoneal space. From initial lateral and abdominal X-ray films, we identified a total of 6 piglets with wires (Table 2). The wire from the 7th piglet likely was lost in the uterine wall during injections; we do not know specifically from which of sow 31-7’s fetuses was lost from. All wires appeared to be in the abdominal cavity of the piglets. In addition, a mummified piglet was found to have a wire, but because its small size makes it difficult to distinguish where the wire was located, the wire location was undetermined. In addition, because we identified all 3 injected animals in the 348 litter by finding the 3 contained wires, mummy 348-05 was not directly due to injection trauma.

After initial X-rays, livers were removed from all piglets and were X-rayed separately for wire detection. We found that one piglet (317-01) had a wire within the liver (Figure 3). All other piglets with a wire did not have a wire in the liver, suggesting that the intraperitoneal space contained wire within the intraperitoneal space. Examples of piglets with wire in liver (317-01) and intraperitoneal space (317-08 and 348-03) are shown in Figure 4. Of the 7 fetuses that were injected with wire, on the basis of observations during surgery, we anticipated that 3 of these fetuses were injected in the liver. Radiographing revealed that only one piglet was successfully injected in the liver, whereas all others with wire were injected intraperitoneally.

Figure 3.

Liver X-rays for wire detection. After initial body X-rays were acquired, livers were removed and radiographed. Only animal 317-01 had wire within the liver.

Figure 4.

Titanium wire detection in the liver and intraperitoneal space of farrowed piglets. These radiographs show that piglet 317-01 had a wire within the liver, whereas piglets 317-08 and 348-03 had wires within the intraperitoneal space.

Discussion

Previous work is reported on the use of laparotomy for injection of human stem cells and liver cells into fetal piglets.^3,4,10,13,14 However, a comprehensive and detailed account of surgical procedures, postoperative care, as well as FL and intraperitoneal space aiming has not been previously published. Here we report the successful injection of piglets within the intraperitoneal space after laparotomy procedures on 2 pregnant gilts.

Laparotomy is an alternative to abdominal injection. There are 2 primary methods for the delivery of cells into fetal piglets. One method involves the insertion of a needle through the abdominal wall and uterus,¹ while the other involves laparotomy procedures to directly inject through the uterine wall.^3,10,13,14 Although injection through the skin is less invasive to the gilt, manipulation of the uterus aids in aiming for specific locations within the fetus. In addition, this manipulation when injecting through the uterine wall helps to avoid vital fetal organs during injection. Furthermore, injection through the dam's skin would require the use of a larger gauge needle, which may cause greater trauma to the fetuses; a smaller gauge needle can be used in laparotomies.

These laparotomy procedures can be accomplished by making a paramammary incision. We chose to perform our laparotomies through a ventral midline incision for a variety of reasons. Such incision location allows for better exposure to both horns of the uterus, as compared with the paramammary approach, in which incision would be closer to one horn than the other. In addition, the midline incision goes through the linea alba, which can promote better healing as compared with going through muscle tissue. By covering the incision with the Ioban bandage and housing the animals on solid flooring without bedding, the increased risk of contamination with a midline incision is mitigated. Because gilts were used for the surgery, the mammary tissue that may present a problem for midline incisions in sows was not encountered.

Wires in the abdominal cavity can be detected by using several methods. In the current study, it was important to establish procedures for wire injections so that the injection site could be determined. We designed a copper wire plunger system for our fetal injections that was used in a 22-gauge needle with a 0.25-mm diameter titanium wire. A 22-gauge spinal needle with a stylet could also be used for titanium wire injections.

In our study, we radiographed piglets after they were born rather than radiographing the pregnant gilts immediately after surgery. There may be limitations in radiographing gilts or sows after surgery, depending on equipment available at different facilities. In addition, we decided to radiograph piglets after birth because doing so mimics the procedure that would be followed if piglets were injected with cells. Between surgery and the end of gestation, a total of approximately 75 d elapsed, so it is possible that wires could move to a different site in the fetus. Radiography or CT imaging could be performed on a gilt or sow after surgery, but matching fetuses seen in utero with farrowed piglets may be difficult. When C-sections are performed, the positions of piglets can be matched with the locations of the injected fetuses.

Regarding the prevention of abortion and death of fetal pigs, altrenogest (Matrix) has historically been used to synchronize estrous in swine. During pregnancy, the body naturally produces progesterone. Altrenogest is a synthetic progestin and therefore can be used to help maintain pregnancies that might otherwise have been aborted due to the stress of the procedure. Altrenogest has been used as a treatment in dolphins¹⁶ and mares¹² to prevent abortions.

During surgery, it is important to minimize manipulation of the fetuses and uterus. In addition, numerous needle punctures should be avoided on a single fetus, to reduce the likelihood of blunt trauma to any internal organs. Together, these practices can help reduce fetal and uterine stress. We demonstrated herein that a single injection from a 22-gauge needle caused minimal trauma to the fetuses, given that 5 of the 6 piglets shown to carry titanium wire were fully developed at the end of gestation.

Overall, we aimed to develop procedures that could be used for cell injections into fetal piglets. As more swine models are developed, whether immunocompetent or immunocompromised, they will continue to be used for translational research relevant to human health. One application for the procedures described here is the injection of human HSC into SCID pigs for the development of a humanized model. To accomplish this task, protocols must be developed that ensure the viability of all piglets and pregnant sows during and after surgical procedures. The descriptions outlined here can be used as guidelines for future research requiring laparotomies and fetal injections in swine.