Outcomes following combined median sternotomy and ventral midline laparotomy for bicavitary penetrating wooden foreign bodies

Abstract

Combined abdominal and thoracic pathology caused by extra-gastrointestinal migration of an ingested wooden foreign body (WFB) is an uncommon but serious injury. Presenting clinical signs are typically nonspecific and, in the absence of observed WFB ingestion, diagnosis is challenging. Treatment requires concurrent abdominal and thoracic surgical exploration to remove the WFB and address injuries caused by its migration. This case series describes perioperative characteristics and outcomes in 4 dogs following combined median sternotomy and ventral midline laparotomy (CMSVML) for bicavitary penetrating WFBs.

Key clinical message:

Treatment of bicavitary penetrating WFBs with CMSVML provided postoperative outcomes similar to those in previous reports; however, high-grade complications and prolonged hospitalization were commonly encountered.

Penetrating wooden foreign bodies (WFBs) are usually encountered as oropharyngeal injuries in dogs (1–3). However, intracavitary WFBs from external penetration or gastrointestinal (GI) tract migration have been reported with injuries to the heart, vasculature, lungs, liver, GI tract, urinary tract, and spinal cord (4–10). In addition to the risk of fragmentation and bacterial inoculation, intracavitary WFBs have the potential to migrate, causing further tissue damage, which demonstrates the importance of prompt diagnosis and treatment (6,11,12). Ingested WFBs have potential to traumatize both the thoracic and abdominal cavities after extraluminal GI migration, requiring surgical exploration of both cavities concurrently (6,7,10,11).

Diagnosing penetrating WFBs can be difficult, especially when ingestion of the WFB is not observed. Presurgical recognition of the extent of injuries and surgical intervention required is critical to successful management of these cases (6,13,14). Orthogonal radiographs have low sensitivity in detecting WFBs (2,13). Compared to other modalities, computed tomography (CT) has been reported as the most accurate imaging modality for detecting WFBs (2,7,11,13,15).

In humans requiring combined thoracotomy and laparotomy for penetrating thoracoabdominal injuries, a higher mortality rate has been reported when compared to single-cavity injuries (15). This is likely due more to the severity of injuries sustained than to the required dual-cavity surgical approach. Survival-to-discharge rates have been reported to be high, but prolonged hospitalization is common in both humans and dogs (6,11,16–20). The purpose of this case series is to describe perioperative characteristics and outcomes in 4 dogs following combined median sternotomy and ventral midline laparotomy (CMSVML) for ingested bicavitary penetrating WFBs.

Case descriptions

Medical records of dogs diagnosed with penetrating WFBs that underwent CMSVML at the Ontario Veterinary College (Guelph, Ontario) between 2014 and 2021 were retrospectively reviewed. Bicavitary penetrating injury was defined as any tissue damage in both the thorax and abdomen and connected by a diaphragmatic defect caused by penetrating WFBs identified on CT imaging and/or at the time of surgery. Four dogs [2 males, 2 females; median age: 2.5 y (range: 0.5 to 11 y); median weight: 19.4 kg (range: 5.5 to 25.2 kg)] met inclusion criteria. Breeds were 1 each of labradoodle, west Highland white terrier, husky, and Alaskan malamute. Median time-to-presentation was 4 d (range: 3 to 21 d). Two dogs were historically observed to consume wooden skewers. Clinical signs and physical examination findings included lethargy (n = 4), hyporexia or anorexia (n = 3), tachypnea (n = 3), increased respiratory effort (n = 3), abdominal pain (n = 2), vomiting (n = 3), and pyrexia (n = 2). Survey thoracic or abdominal radiographs did not identify a WFB in any dog; however, CT under general anesthesia identified a WFB in 3 dogs (Figure 1). In Dog 3, a WFB was not identified on CT, but concurrent imaging findings, including pulmonary heterogenicity and suspect visceral pleural damage, marked cellulitis and pleuritis at the level of the 8th rib, and a linear tract in the right lateral liver lobe with small adjacent peritoneal gas bubbles, supported the presence of a migrating WFB.

Figure 1.

Transverse plane (A to D) and dorsal plane (E) computed tomographic images of Dog 1, showing a wooden foreign body (arrowhead) extending from the right caudal pleural space into the abdomen and stomach. There is moderate pleural effusion (star) in the dorsal pleural space.

All dogs were taken for exploratory surgery where a CMSVML was done, with ventral midline laparotomy completed prior to median sternotomy in 3 dogs. In Dog 3, a median sternotomy was done first due to severe pneumothorax and respiratory decompensation. In Dog 1, thoracoscopic evaluation was completed prior to median sternotomy. A WFB was identified and removed in all cases (Figure 2). Surgical procedures are summarized for each dog (Table 1). Multiple lung lobectomies were required in all dogs. Partial and/or complete lung lobectomy was done using a TA30V (Medtronic, Minneapolis, Minnesota, USA) or Endo GIA (Medtronic) stapling device. The thoracic and abdominal cavities were flushed copiously with sterile saline prior to placement of unilateral (n = 3) or bilateral (n = 1) thoracic drainage catheters (MILA International, Florence, Kentucky, USA) and closed-suction abdominal drains (n = 2) (Cardinal Health, Dublin, Ohio, USA). Dog 4 required repositioning for exploration and debridement of a right lateral thoracic granuloma that communicated with the thoracic cavity, suspected to be due to migration of the WFB. Median surgical anesthesia time was divided, as 3 dogs underwent a separate general anesthesia for CT imaging on the day before surgery [median: 225 min (range: 195 to 285 min)], whereas 1 dog went straight from CT to surgery (315 min).

Figure 2.

Intraoperative image from a median sternotomy in Dog 2, showing a wooden skewer (arrows) penetrating through the stomach (middle arrow) and diaphragam (top arrow). This wooden skewer was found to have injured the left caudal and caudal subsegment of the left cranial lung lobe.

Table 1.

Origin and location of wooden foreign body (WFB), surgical procedures completed, and outcome in each dog undergoing combined median sternotomy and ventral midline laparotomy.

Dog	Origin & location of WFB	Diaphragmatic herniorrhaphy	Thoracic surgical procedures	Abdominal surgical procedures	Survived
1	Gastric: Extending ~16 cm cranially from the gastric lumen to right caudal lung lobe.	No	Complete lobectomy of the right cranial and right middle lung lobes. Partial lobectomy of the right caudal lung lobe.	Gastric perforation repair.	Yes
2	Gastric: Extending ~11.5 cm cranially from the gastric lumen to the left cranial lung lobe.	Yes	Partial lung lobectomy of the caudal subsegment of the left cranial lung lobe. Complete left caudal lung lobectomy.	Gastric perforation repair. L3/L4 epaxial abscess drainage and exploration.	Yes
3	Gastric: Extending cranially from the gastric lumen to the left cranial lung lobe.	Yes	Complete lung lobectomy of the left caudal, accessory, and right middle lobes. Partial lung lobectomy of the caudal subsegment of the left cranial lobe.	Gastric perforation repair.	No
4	Gastric: Extending ~15.5 cm cranially and dorsolaterally from the gastric lumen to the subcutaneous tissue at the level of the 11th rib.	Yes	Complete lung lobectomy of the left caudal lung lobe. Lateral thoracic granuloma debridement.	Gastric perforation repair.	Yes

Intraoperative, early postoperative (before discharge), and late postoperative (after discharge) complications were graded on a scale of 1 (mild clinical signs) to 5 (death) using the Veterinary Cooperative Oncology Group — Common Terminology Criteria for Adverse Events (VCOG-CTCAE v2) (21). Long-term outcomes were obtained via an online survey sent to referring veterinarians. Complications encountered are outlined in Table 2. A pneumothorax (grade 4) developed in Dog 3 in the early postoperative period, requiring continuous pleural suction and positive pressure ventilation. This dog also experienced dyspnea and hypoxemia upon anesthetic recovery, precluding extubation and requiring continuous positive pressure ventilation and pleural suction. This dog also had severe hypovolemia intractable to vasopressor support and plasma transfusion. Euthanasia was elected ~6 h after surgery due to a poor prognosis. The remaining 3 dogs were discharged from hospital with a median time-to-discharge of 6 d (range: 4 to 9 d). Long-term follow-up was available in 2 dogs (at 16 and 20 mo, respectively). Both were alive at the time of writing, and full recovery and excellent owner satisfaction were reported. One dog was lost to follow-up after the 14-day suture removal appointment.

Table 2.

Types, frequencies, and grades of Veterinary Cooperative Oncology Group — Common Terminology Criteria for Adverse Events (VCOG-CTCAE v2) complications encountered in the intraoperative, early postoperative (prior to discharge), and late postoperative (after discharge) periods.

Interval	VCOG-CTCAE v2 complication grade
	1	2	3	4	5
Intraoperative complications	Regurgitation (1)	Hypotension (2) Hypoxia (1)	Hypotension (2) Hypothermia (2)
Early postoperative complications	Regurgitation (1)			Hypovolemia (1) Dyspnea (1) Hypoxemia (1) Pneumothorax (1)
Late postoperative complications		Surgical site infection (1)

Discussion

Penetrating WFBs represent challenging cases because of the potential for bicavitary injuries that often involve multiple organs or tissues, and they can be unpredictable given the possibility for WFB migration (4–10). In this case series, CMSVML allowed for detailed exploration of thoracoabdominal cavities as well as debridement and repair of traumatized tissues. Although CMSVML resulted in prolonged postoperative hospitalization, dogs can have excellent long-term outcomes. This was consistent with results in humans, where CMSVML was a safe and effective surgical approach for management of thoracoabdominal pathology (17,18).

In the present case series, 3 dogs underwent ventral midline laparotomy prior to median sternotomy. However, median sternotomy was prioritized in Dog 3 due to a severe pneumothorax and respiratory decompensation. This dog was euthanized in the early postoperative period. Mortality does not appear to have been related to inappropriate surgical sequencing, as only minor injuries were noted in the abdomen. Published data on the sequence of surgical procedures for penetrating thoracoabdominal injuries in humans showed inappropriate sequencing, defined as the need to interrupt one procedure and convert to another, in nearly 1/2 of patients (16). However, a standardized sequencing recommendation would be inappropriate in dogs with bicavitary WFBs, as the injuries and clinical status of the individual dog can be highly variable, as in this and other previous reports (4–10).

Lung lobectomy using a stapling device was the most common surgical procedure in this case series. Stapling devices are a secure and efficient technique for complete and partial lung lobectomies, with low reported complication rates in both veterinary and human medicine (22,23). Dog 3 in this case series developed a persistent pneumothorax postoperatively (22,24). It is unknown if this complication was due to failure of a stapling device in occluding a bronchus, a missed laceration, or a missed abscess in one of the remaining lung lobes. This dog required extensive lung lobectomies approximating at least 50% of total lung volume due to extensive abscessation and lacerations (Table 1). Although not lateralized, this degree of lung volume resection is similar to a pneumonectomy, which has been reported to carry high morbidity and mortality rates in dogs (25,26). Right pneumonectomy, accounting for 58% of lung volume, can be tolerated in healthy dogs and those with chronic progressive disease, as the remaining lung volume can compensate (25–28). In Dog 3, insufficient residual healthy lung tissue to compensate for the removed lung volume likely led to respiratory failure and subsequent euthanasia.

Survey thoracic or abdominal radiography, a common first-line diagnostic approach for primary or urgent care veterinarians, has been reported to have a very low positive predictive value in identifying WFBs (13,14). None of the WFBs in our case series were identified on radiographs. In comparison, CT has been reported as the most accurate modality for WFB detection, with a sensitivity and specificity of 79 and 93%, respectively (2,12,15). However, varying wood densities and the degree of body fluid absorption once in situ result in a wide range of possible attenuation on CT, which can make identification challenging (9,12). In addition, CT findings of acute WFBs often mimic gas due to their porous composition, whereas chronic, hydrated WFBs appear similar to soft tissue (7,9,15). The hallmark finding in acute cases tends to be soft tissue emphysema, whereas chronic cases present with draining sinuses, cavitary lesions, fat stranding, and periosteal reactions on adjacent bones (12). Furthermore, WFBs are identified on CT imaging more commonly in chronic cases in both veterinary and human medicine (9,12).

The VCOG-CTCAE v2 scheme was chosen to report both intraoperative and postoperative complications. A need for standardized adverse event (AE) reporting in veterinary medicine, to allow meaningful comparison across studies, has been identified (29). The VCOG-CTCAE v2 scheme was chosen as it was formulated to provide standardized definitions and specific grading criteria for individual AEs, reducing the extent of subjective judgements (19).

Long-term survival rates for dogs with bicavitary WFBs who underwent CMSVML are limited to a subset population of 11 dogs in 3 retrospective studies. Ten dogs survived to discharge (6,10,11). In our case series, similar outcomes were observed. Two dogs with long-term follow-up were reported to achieve excellent outcomes.

Limitations of this study include its retrospective nature, small sample size, and selection bias. Data were retrieved from medical records and multiple clinicians were involved in the management of these cases.

In conclusion, acceptable outcomes can be achieved following CMSVML for bicavitary penetrating WFBs. As considerable morbidity is expected, these cases may be best managed at referral hospitals with specialist care.