The Case: Common Surgery Gone Awry

History/Presentation

A 5-year-old neutered male Jack Russell terrier was presented 3 days postoperatively with complaints of lethargy, abdominal pain, and inappetence. An enterotomy had been performed for a distal jejunal foreign body (grass). The patient had been treated the day of enterotomy with IV fluids, cefazolin, morphine, and meloxicam and discharged with amoxicillin/clavulanic acid, firocoxib, and tramadol.

Physical Examination

• Temperature: 100.5°F (38.5°C)

• Heart rate: 160 bpm

• Respiratory rate: 60 bpm 

• Mucous membranes: pale pink

• Body weight/body condition score: 9.22 kg/6 of 9

• Midline abdominal incision was clean, dry, and intact with no discharge

• Cardiopulmonary: no murmurs or arrhythmias, pulses fair/synchronous, normal bronchovesicular sounds

• Normal ambulation

• Abdominal palpation: very tense, moderately painful. Firm mass effect in mid-cranial abdomen. Patient splinted on palpation in this region.

Diagnostic Findings

• aFAST* scan: ascites present

• Complete abdominal ultrasound: decreased GI motility and abnormal-appearing mesentery and small intestine with focal region of fluid pocketing in mid-abdomen. The fluid had increased echogenicity. High degree of suspicion was raised for septic peritonitis. 

• Abdominocentesis: 

  • In-house cytology: large number of degenerate neutrophils with intracellular cocci and rods 

  • Submitted ascitic fluid sample for aerobic/anaerobic culture

• Blood pressure: 90 mm Hg, systolic; 1-hour recheck, 105/68

Blood analysis

• Packed cell volume: 42% (range, 36%-60%), serum slightly hemolyzed

• Total protein: 5.6 g/dL (range, 5.0-7.4) (56 g/L [range, 50-74])

• White blood cells: 13,130/μL (range, 4000-15,500) (13.1 x 10⁹/L [range, 4.0-15.5])

  • Neutrophils: 10,960/μL (range, 2060-10,600) (11.0 x 10⁹/L [range, 2.1-10.6])

• Platelets: 83,000/μL, with clumping (range, 170,000-400,000) (83.0 x 10⁹/L [range, 170-400])

• Sodium: 134.8 mEq/L (range, 139-154) (134.8 mmol/L [range, 139-154])

• Potassium: 5.49 mEq/L (range, 3.6-5.5) (5.49 mmol/L [range, 3.6-5.5])

• Glucose: 102 mg/dL (range, 70-138) (5.7 mmol/L [range, 3.9-7.7])

• Blood urea nitrogen: 37 mg/dL (range, 6-31) (13.2 mmol/L [range, 2.1-11.1])

• Lactate: 1.89 mmol/L (range, 0.6-2.9)

• Creatinine: 1.0 mg/dL (range, 0.5-1.6) (88 μmol/L [range, 44-141])

• Alanine aminotransferase (ALT): 568 IU/L (range, 12-118)

• Alkaline phosphatase (ALP): 1021 IU/L (range, 5-131)

• Total bilirubin: 2.1 mg/dL (range, 0.1-0.3) (36 μmol/L [range, 1.7-5.1])

• Albumin: 1.7 g/dL (range, 2.7-4.4) (17 g/L [range, 27-44])

Paired lactate levels

• Blood: 1.89 mmol/L

• Abdominal effusion: >12.0 mmol/L

Paired glucose levels

• Blood: 120 mg/dL (6.7 mmol/L)

• Abdominal effusion: 26 mg/dL (1.4 mmol/L)

Treatment

Exploratory laparotomy was performed at the previous jejunal enterotomy site. Dehiscence and mild erythema of the surrounding intestines were noted. A resection and anastomosis were performed. Patient recovered uneventfully from anesthesia. 

• Lactated Ringer’s solution: 20 mL/kg bolus IV; continued at 13.5 mL/kg/hr for 3 hours

• Fentanyl: 2 mcg/kg/hr

• Dolasetron: 0.6 mg/kg q24h IV 

• Famotidine: 0.5 mg/kg q24h IV

• Ampicillin/sulbactam: 22 mg/kg q8h IV

• Enrofloxacin: 10 mg/kg q24h IV

• Based on Postoperative Diagnostic Findings (below):

  • Hydroxyethyl starch colloid: 3 mL/kg bolus; then 0.6 mL/kg/hr 

  • Lactated Ringer’s solution: 5.4 mL/kg/hr

Postoperative Diagnostic Findings

• Packed cell volume: 32% (range, 38%-60%)

• Total protein: 4.0 g/dL (range, 5.0-7.4) (40 g/L [range, 50-7])

• Blood pressure: 90 mm Hg, systolic

Blood pressure 1 hour later was 150/68.

DAY 2

Diagnostic Findings

• Blood analysis

  • ALT: progressive elevation

  • ALP: progressive elevation

  • Albumin: 1.3 g/dL (range, 2.7-4.4) (13 g/L [range, 27-44])

  • Globulin: 1.4 g/dL (range, 1.6-3.6) (14 g/L [range, 16-36])

  • Packed cell volume: 20% (range, 36%-60%)

  • White blood cells: 26,000/μL (range, 4000-15,500) (26 x 10⁹/L [range, 4.0-15.5])

    • Bands: 4,300/μL (range, 0-300) (4.3 x 10⁹/L [range, 0-0.3])

  • Platelets: 49,000/μL, with clumping (range, 170,000-400,000) (49 x 10⁹/L [range, 170-400])

• Prothrombin time (PT)/partial thromboplastin time (PTT): 9.7 sec (range, 5.4-8.8)/32.0 sec (range, 9.1-15.6)

• Colloid oncotic pressure: 9.1 mm Hg (range, 18-24)

• Blood pressure: within normal limits, serial Doppler readings

Treatment

A central line was placed.

• Packed red blood cells: 130 mL over 6 hours

• Human serum albumin: 25% of 100 mL over 24 hours 

• Hydroxyethyl starch 6% colloid: 25 mL/kg/day

• Plasmalyte†: 10 mL/kg/day 

• Procalamine‡:  40 mL/kg/day

• Fentanyl (4 mcg/kg/hr) / lidocaine (1 mg/kg/hr)/ketamine (4 mcg/kg/min) CRI 

• Pantoprazole: 1 mg/kg q24h IV

• Misoprostol: 3 mcg/kg q8h PO

• Ranitidine: 2 mg/kg q12h IV

• S-Adenosylmethionine/silybin: 225 mg q24h PO

• Vitamin K₁: 1 mg/kg q24h SC

• Ampicillin/sulbactam: 22 mg/kg q8hr IV

• Enrofloxacin: 10 mg/kg q24h IV

• Maropitant: 1 mg/kg q24h SC

• Dolasetron: 0.6 mg/kg q24h IV

DAY 3

Patient had developed abdominal distension presumed to be ascites. A Jackson-Pratt drain had been recommended at the time of corrective surgery, but the surgeon felt that the drain was not indicated based on the overall appearance of the abdomen.

Diagnostic Findings 

• Thoracic/abdominal radiographs: pulmonary infiltrates compatible with pneumonia, abdominal distension presumed to be ascites

• Total protein: 3.5 g/dL (range, 5.0-7.4) (35 g/L [range, 50-74])

• Blood urea nitrogen: 45 mg/dL (range, 6-31) (16.1 mmol/L [range, 2.1-11.1])

• Glucose: 106 mg/dL (range, 70-138) (5.9 mmol/L [range, 3.9-7.7])

• ALT: 1679 IU/L (range, 12-118)

• Aspartate aminotransferase: 1423 IU/L (range, 13-81)

• ALP: 2572 IU/L (range, 5-131)

• Gamma-glutamyl transferase (IU/L): 1.0 (range, 0-25)

• Bilirubin (mg/dL): direct–7; indirect–7.3; total–45 (range, 0.1-0.3)

  • (Bilirubin [μmol/L]: direct – 120; indirect – 125; total – 770 [range, 1.7-5.1])

• Albumin: 2.7 g/dL (range, 2.7-4.4) (27 g/L [range, 27-44])

• Globulin: 0.8 g/dL (range, 1.6-3.6) (8 g/L [range, 16-36])

• Packed cell volume: 22% (range, 36%-60%)

• White blood cells: 51,700/μL (range, 4000-15,500) (51.7 x 10⁹/L [range, 4.0-15.5])

  • Neutrophils: 42,394/μL (range, 2060-10,600) (42.3 x 10⁹/L [range, 2.1-10.6])

  • Bands: 5,887/μL (range, 0-300) (5.9 x 10⁹/L [range, 0-0.3])

• Platelets: 36,000/μL (range, 170,000-400,000) (36 x 10⁹/L [range, 170-400])

• PT/PTT: 15.5 sec (range, 5.4-8.8)/>100 sec (range, 9.1-15.6)

• Fibrinogen: <50 mg/dL (range, 150-300) (<1.5 μmol/L [range, 4.4-8.8])

• D-dimer: <250 mcg/mL (range,  0.0-0.4) (<1369 nmol/L [range, 0-2.2])

Outcome

Euthanasia was elected due to poor prognosis. 

*Abdominal focused assessment with sonography in trauma

†A multiple electrolyte solution

‡3% amino acid and 3% glycerin injection with electrolytes

The Generalist’s Opinion

Barak Benaryeh, DVM, DABVP

Intestinal enterotomies are performed in nearly all veterinary practices. This case is a reminder of the complexity of the procedure and its potential complications. Studies have documented that postenterotomy dehiscence rates range from as low as 2%¹ to as high as 15.7%.²  It is important to safeguard against this complication and, should it occur, be able to recognize and treat it appropriately.

Playing It Safe

The patient in this case was treated with meloxicam on the day of surgery and then discharged with firocoxib. Treatment with nonsteroidal anti-inflammatory drugs (NSAIDs) can lead to gastrointestinal adverse effects, and mixing different NSAIDs, as was done in this case, increases the risks. The half-life of meloxicam is approximately 24 hours, so the drug would still have been active when the firocoxib was initiated. A washout period is recommended when changing from one NSAID to another. While there are differences in opinion as to how long a washout period should be, it is generally agreed that some washout period is needed. This dog had none, despite having a known gastrointestinal issue. It is safer to allow for possible mucosal compromise in the gastrointestinal tract by limiting any use of NSAIDs. There is no telling how much of a role, if any, NSAIDs played in this dog’s illness. One should, however, always err on the side of safety when using NSAIDs in surgery involving the GI tract.

Recognizing a Problem

The clinicians in this case did an excellent and thorough job of recognizing the septic peritonitis.  Paired lactate and paired glucose levels were a diagnostic aid, but noting intracellular bacteria in the abdominal fluid was diagnostic.³

Surgical Treatment

Septic peritonitis requires surgery. The veterinarians made the appropriate decision to go back in to perform a resection and anastomosis. Once the correction has been made, the postoperative choices are to close the abdomen, leave an open abdomen, or place a closed suction drain such as a Jackson Pratt. Leaving an abdomen to drain via either open or closed-suction drainage requires a level of familiarity with the technique as well as significant aftercare and may be beyond the comfort zone of many general practitioners. Studies comparing outcomes between closing an abdomen, open peritoneal drainage, and closed-suction drains have been inconclusive,³ but it is difficult to compare cases as there are many limitations to any study with regard to success rates. It is ultimately up to the surgeon, including his or her level of comfort with each technique and how effective he or she feels the removal of contaminated material was.

Medical Treatment

In addition to addressing surgical concerns, patients with septic peritonitis need to be stabilized medically. This dog was treated with crystalloid therapy to replace fluid losses. Colloid therapy was also started to increase the low oncotic pressure caused by the hypoproteinemia commonly seen in patients with peritonitis. Increased vascular permeability causes fluid and albumin to leak into the peritoneal cavity. On day 2 this dog’s proteins, packed cell volume, and colloid oncotic pressure were dropping. In addition, the PT and PTT were increasing. One additional consideration would have been the administration of fresh frozen plasma, which would have helped restore colloid oncotic pressure as well as replenish clotting factors. Given, however, that the patient progressed to what was likely disseminated intravascular coagulation (DIC), it is unlikely that anything would have saved this dog. Mortality rates between 50% and 68% have been reported in septic peritonitis cases.^3,4 

The important point to remember here is how dire the consequences are of a dehiscence in the gastrointestinal tract. Practitioners must remember to take every precaution possible to prevent this complication: leak test all your surgical sites, wrap omentum around the intestinal closure, and err on the conservative side with any postoperative drug prescriptions.

The Specialist’s Opinion

Karl Maritato, DVM, DACVS (Small Animal)

This dog’s euthanasia is an example of how even what are presumed routine cases can have poor outcomes. While the initial treatment 3 days prior to presentation appears relatively appropriate, a few valid questions can be raised. Likely, the most pressing is why the patient experienced dehiscence of the initial enterotomy. Dehiscence of an enterotomy for foreign bodies occurs in 10% to 15% of patients.¹ Among its causes are poor quality of the bowel, impaired blood flow to the region, low albumin, and improper tissue handling.¹ It is hard to speculate as to the cause in this particular case as the information on these aspects is not offered.

Nonsteroidal Anti-inflammatory Agents (NSAIDs)  

The patient was given 2 different NSAIDs after surgery. While mixing NSAIDs is strongly discouraged, there is currently no scientific evidence in the veterinary literature that suggests using a single NSAID would lead to an increased risk for dehiscence of an enterotomy. In human medicine, the literature currently is contradictory.^2,3 In most surgeons’ hands, however, NSAID use after intestinal surgery is not commonplace, due to the associated risks for gastrointestinal complications.

The Corrective Surgery

On initial presentation 3 days after enterotomy, the patient workup appears adequate, resulting in proper surgical exploration. Resection of the dehisced enterotomy site is the indicated procedure, followed by thorough abdominal lavage. A common question is whether primary closure is sufficient or either closed drainage or an open abdomen should be employed. The literature is also contradictory on this point, with no single technique proven superior.^4-7 An open abdomen is racked with complications and does not accrue any benefit over closed drainage; therefore, this author does not recommend that technique. Because minimal morbidity is associated with a closed suction drain, it is best to err on the side of caution and place a drain, although it is not an absolute. Fluid production after peritonitis depends on the amount of peritoneal inflammation associated with the initial cause, and it is much more difficult for intestinal incisions to heal in the face of abdominal fluid due to its effect on the inflammatory/lag phases of healing. Mild, focal peritonitis frequently does not produce much fluid and does not require a drain. More diffuse inflammation leads to voluminous fluid production.  

The bloodwork abnormalities (elevated bilirubin, ALT, and ALP) at the time of presentation suggest sepsis and thus proper antibiotic coverage is critical (which was implemented). It would have been reasonable to check the PT/PTT to get a baseline prior to surgery. This factor was checked on the second postoperative day and clotting time was beginning to elevate. This development, combined with low platelets on day 2, is pathognomonic for DIC. Fresh frozen plasma is the treatment of choice for affected patients, as it is able to supply oncotic pressure and all clotting factors except von Willebrand. The DIC was likely caused by the progressive septicemia, as evident from the continued elevation of bilirubin and liver enzyme values, along with reductions in albumin and white blood cells. 

Loss of albumin into the forming abdominal effusion is also a cause of hypoalbuminemia. Hypoalbuminemia as a cause of intestinal dehiscence is also controversial, as there are contradictory study results.⁸ However, hypoalbuminemia leads to interstitial edema, which does impede healing, and also slows blood flow at the capillary beds, which can worsen the effects of hypotension and septicemia as well as increase lactate production, leading to acidosis.  

On the day of euthanasia, there was an increase in abdominal fluid. What we do not know is whether this fluid was hypoproteinemic ascites or recurrent septic peritonitis, because we do not have documentation of abdominocentesis with cytology or paired glucose tests.  However, there is evidence of worsening DIC and sepsis despite aggressive management.

The survival rate for septic peritonitis ranges from 30% to 70%, with an average of 50%.¹ The overall management of this patient pre- and postoperatively was aggressive and proper. While it is possible to question some of the decisions made in the treatment, the literature is not in agreement on many of the questions that would be posed, so it is difficult to make absolute recommendations as to what (if anything) could have changed the outcome in this case.