Pneumatosis Coli in a Dog
ABSTRACT
The objective of this report is to document a case of pneumatosis coli in an immunosuppressed dog successfully treated with supportive care that included hyperbaric oxygen. A 9 yr old, neutered male Maltese was presented for hematochezia of ∼36 h duration. The dog had a history of immune-mediated thrombocytopenia that was well managed with prednisone and azathioprine. Initial evaluation was unremarkable except for increased rectal temperature and hematochezia on digital rectal examination. Abdominal radiographs revealed diffuse gas within the colonic wall consistent with pneumatosis coli. Scant-free air was also appreciated in the peritoneal and retroperitoneal cavities. The dog was managed for 4 days in hospital with broad spectrum antibiotic, anthelminthic, and gastroprotectant therapy. He received one session of hyperbaric oxygen therapy while in hospital. Repeat abdominal radiographs after 72 hr of medical management showed improvement of the pneumatosis coli. After 4 days of hospitalization, he was discharged with resolution of clinical signs. Pneumatosis coli is rarely diagnosed in dogs but should be considered in cases with clinical signs of colorectal disease.
Introduction
Pneumatosis coli is rarely diagnosed in veterinary patients, and multiple theories exist as to its etiology. Gas containing cysts form in the subserosal and submucosal layers of bowel due to abnormal mucosal permeability or pressures.1,2 A broad spectrum of disease has been described in human literature, from a benign, self-resolving form to a pathologic form that can lead to viscus rupture and septic peritonitis.3 The incidence of pneumatosis coli has not been described in veterinary patients. The condition has been associated with immunosuppression in humans; however, the mechanism of that relationship is unknown.4–6
The mechanical theory of pneumatosis coli describes a defect in the intestinal mucosa, such as an ulcer, or an increase in intraluminal pressure that allows gas to pass into the bowel wall. This can occur secondary to obstructive disease or iatrogenically after endoscopy.2 Gas enters the colon wall from within the bowel lumen and is distributed along the length of bowel by peristalsis. Although pneumatosis coli can occur with an increase in intraluminal pressure alone, frequently, there is concurrent mucosal injury.7 Mechanical pneumatosis coli is more often associated with underlying gastrointestinal necrosis or inflammatory disease.8,9 Primary gastrointestinal disease should be suspected in cases of pneumatosis coli without evidence of other causes.
Gas-producing bacterial infection of the colonic mucosa has been indicated in cases of pneumatosis coli in both humans and dogs.1,2 It is unclear if the gas in the colon wall is caused by direct bacterial invasion, or if colonic bacteria produce hydrogen at high concentrations, causing a diffusion gradient of hydrogen toward submucosal vessels. The gas dissects into the gut wall and forms pneumocysts.7 Clostridial species induce pneumatosis intestinalis when experimentally inoculated into the intestinal submucosa of guinea pigs; however, Escherichia coli and Enterobacter aerogenes do not.10 In human literature and the few veterinary case reports describing management of this condition, antibiotic therapy is recommended due to the concern for bacterial etiology.11,12 The use of hyperbaric oxygen therapy in pneumatosis coli may be beneficial both for its effect on anaerobic bacteria and dilution of gases within and around the gut wall with oxygen, allowing for easier diffusion of nitrogen out of the pneumocystis.7
In veterinary literature, there is limited information regarding etiology of this disease. Clostridial infection was indicated in one dog, and concurrent hyperadrenocorticism was suspected in a dog for whom no underlying cause was identified.1,12 This report describes a case of pneumatosis coli in a dog receiving long-term prednisonea and azathioprineb therapy.
Case Report
A 9 yr old, neutered male Maltese weighing 7.3 kg was presented for hematochezia of ∼36 hr duration. The dog was diagnosed with immune-mediated thrombocytopenia ∼3 yr prior and was currently well managed on prednisonea (0.3 mg/kg per os [PO] q 48 hr) and azathioprineb (2 mg/kg PO q 48 hr). For 1 yr prior to presentation, the dog had received a home-cooked diet of salmon, white beans, carrots, and brown rice. He had no previous history of gastrointestinal signs. There was no known history of dietary indiscretion or toxin ingestion.
Thirty-six hours prior to presentation, the dog had an episode of loosely formed hematochezia (fecal score two, with one being normal and four being watery with no texture). The dog continued to produce small volumes of mucoid hematochezia every few hours. The dog remained bright and alert with a normal appetite and water intake. No vomiting or regurgitation was observed by the owner. On presentation to the hospital, the dog’s physical examination revealed an increased rectal temperature of 103.1°F, tachycardia at 160 beats per minute, and hematochezia on digital rectal examination. He was nonpainful on abdominal palpation and euhydrated.
Diagnostic Investigation
At presentation, a venous blood gas analysis and abbreviated biochemical profile were performed. These were unremarkable aside from hyperlactatemia of 3.6 mg/dL (reference range 0.4–1.5 mg/dL). A platelet count was 158 K/uL (reference range 134–396 K/uL).
Abdominal radiographs were performed and revealed diffuse, moderate-to-severe gas infiltration of the colon wall with gas (Figures 1A, B). The colonic wall thickening worsened abroad and was measured up to 5 mm at its thickest point. In the caudoventral abdomen and dorsal retroperitoneal space, there were multifocal, ill-defined regions of gas opacity. There was a focal loss of serosal detail in the mid-abdomen, which was concerning for abdominal effusion. These findings were consistent with moderate-to-severe pneumatosis coli and mild pneumohydroperitoneum and pneumoretroperitoneum. To further investigate for an underlying cause of the pneumatosis coli, an abdominal ultrasound was performed. There was a diffusely irregular colonic wall with reverberation artifact, consistent with gas within the colon wall (Figure 1C). Scant peritoneal effusion and focal hyperechoic surrounding mesentery was concerning for focal peritonitis.
Citation: Journal of the American Animal Hospital Association 55, 4; 10.5326/JAAHA-MS-6890
On the morning of the second day of hospitalization, an abdominal-focused assessment with sonography for trauma was performed, and no free peritoneal fluid was seen.13 Blood lactate was rechecked and was within normal limits (1.4 mg/dL, reference range 0.4–1.5 mg/dL).
A complete blood count and biochemical profile on day 4 of hospitalization revealed a mild lymphopenia (0.4 K/uL, reference range 0.9–3.4 K/uL), mild hyperglycemia (141 mg/dL, reference range 79–120 mg/dL), and moderate increase in alkaline phosphatase (513 U/L, reference range 8–114 U/L). Bloodwork abnormalities were most consistent with chronic corticosteroid therapy.14 Repeated abdominal radiographs prior to discharge revealed improvement in the gas colonic infiltration (Figure 2).
Citation: Journal of the American Animal Hospital Association 55, 4; 10.5326/JAAHA-MS-6890
Therapy
Initial medical therapy included metronidazolec (15 mg/kg IV q 12 hr), ampicillin/sulbactamd (45 mg/kg IV q 8 hr), enrofloxacine (15 mg/kg IV q 24 hr), pantoprazolef (1 mg/kg IV q 24 hr), and fenbendazoleg (50 mg/kg PO q 24 hr). Because fecal cultures were not obtained in this dog, four quadrant antibiotic therapy was initiated to treat potential Clostridial infection in the colon, migration of normal colonic flora through the diseased colon wall, and because of concern for sepsis from either of these sources.11 An antiparasitic and proton pump inhibitor were given to address intestinal parasitism and ulcers as potential causes of pneumatosis coli.1,11 Intravenous crystalloid therapyh was initiated at 4 mL/kg/hr. Azathioprineb was continued at 2 mg/kg PO every other day for management of previously diagnosed immune-mediated thrombocytopenia. The dog’s prednisone therapy was discontinued because of concern for gastrointestinal ulceration with an abnormal colonic wall, which would be unlikely with such a low dose and implication of corticosteroid therapy in some human cases of pneumatosis coli.4–6 All of these therapies were continued for the duration of the dog’s hospitalization. For the first 24 hr of hospitalization, solids were withheld PO, then a low-fat, low-fiber dieti was gradually reintroduced.
Hyperbaric oxygen therapy was performed on day 2 of hospitalization. The first session was at two atmospheres absolute for 45 min. The dog vocalized during pressurization of the chamber, but otherwise no adverse effects were reported. The second session on day 3 was stopped prematurely prior to reaching hyperbaric conditions when the dog began chewing out his intravenous catheter.
The dog was hospitalized for a total of 4 days. His hematochezia resolved and he had maintained a normal appetite for the duration of his hospitalization. He was discharged with the following oral medications: amoxicillin-clavulanic acidj (17 mg/kg PO q 12 hr), enrofloxacink (9 mg/kg PO q 24 hr), metronidazolel (10 mg/kg PO q 12 hr), omeprazolem (1.5 mg/kg PO q 24 hr), and fenbendazoleg (50 mg/kg PO q 24 hr, continued for 2 days after discharge). The dosing of azathioprineb (2 mg/kg PO) was increased to once daily because prednisone was discontinued and there was concern for recurrence of his previously diagnosed immune-mediated thrombocytopenia.
One day after discharge, the dog represented to our facility for two episodes of vomiting within 24 hr. The dog’s appetite was slightly decreased that morning. An abdominal-focused assessment with sonography for trauma revealed no free abdominal fluid. Abdominal radiographs were consistent with improving pneumatosis coli and showed no evidence of mechanical obstruction or pneumoperitoneum (Figure 2). The dog was treated as an outpatient with subcutaneous crystalloidsh (27 mL/kg once) and maropitantn (1 mg/kg subcutaneously once).
The dog was re-evaluated 4 days after discharge. He was clinically normal, and his platelet count was within the reference range.
Discussion
This report documents a case of pneumatosis coli in a Maltese who resolved with medical management including hyperbaric oxygen therapy, although the underlying cause was not determined. This case was diagnosed on survey abdominal radiographs. Computed tomography is a more sensitive diagnostic modality for pneumatosis coli and is recommended for diagnosis of this disease in humans.11 Abdominal ultrasound can be used to evaluate for underlying disease, and irregularity of the colon wall is supportive of pneumatosis coli. Further diagnostic tests that could have been performed to evaluate etiopathogenesis of this dog’s pneumatosis coli include fecal floatation, fecal culture, and colonoscopy with biopsy and tissue culture.
This dog was previously diagnosed with immune-mediated thrombocytopenia and had been receiving combination therapy of prednisone and azathioprine for ∼8 mo prior to presentation to our institution. Pneumatosis coli in people has been associated with corticosteroid administration and immunosuppression.15 Compromise of the intramural lymphoid tissue and inhibition of tissue repair, leading to reduction of normal luminal barrier function may predispose to bacterial intrusion and gas diffusion into the gut wall.7,16 Steroids and other immunosuppressive agents can contribute to mucosal break down in the gastrointestinal tract and predispose to air dissection.15 Chronic immunosuppressive therapy may have contributed to the development of pneumatosis coli in this dog.
There is a correlation with bowel necrosis and increased mortality in people with pneumatosis coli who have a lactate level of > 2 mg/dL and evidence of peritonitis.3,11 In this case, lactate on presentation was 3.6 mg/dL (reference range 0.4–1.5 mg/dL), evidence of focal peritonitis was seen on abdominal ultrasound, and the dog was mildly febrile. A recently published study in people recommends surgical exploration to be strongly considered in patients with increased lactate and/or peritonitis, as they are predictive of pathologic forms of the condition.3 Previous studies have correlated fever, blood per rectum, and free gas with need for surgical intervention in people.17,18 There is no published information regarding prognostic factors in dogs with pneumatosis coli, and the spectrum of disease in dogs is unknown. This case is consistent with other reports of pneumatosis coli in dogs, and clinical signs resolved with medical management within 1 wk of diagnosis.12 As some benign forms of this disease are self-resolving in people, it is unknown if this dog would have recovered fully without medical intervention. The clinical resolution of this case is most consistent with a benign form of the condition, despite increased lactate, focal peritonitis, and increased rectal temperature on presentation. Further investigation is needed to determine if the same predictive variables used in humans can be applied to dogs with pneumatosis coli. To the authors’ knowledge, a pathologic form of pneumatosis coli leading to viscus rupture has not yet been described in veterinary literature.
Hyperbaric oxygen therapy has been recommended for treatment of pneumatosis coli in humans.11 The high oxygen tensions are toxic to anaerobic bacteria, bactericidal to most Clostridial species, and may enhance leukocyte phagocytic activity.11,19 Additionally, high-arterial oxygen tensions promote diffusion of oxygen into the hydrogen-rich pneumocysts. The hydrogen is diluted, and the cysts regress as the oxygen is consumed by normal cellular metabolism.11 In human medicine, there are no randomized, controlled clinical trials using hyperbaric oxygen therapy for treatment of pneumatosis coli due to the small number of cases and potential severity of the disease. However, upon reviewing published cases using hyperbaric oxygen therapy for pneumatosis coli, there were no reported complications associated the intervention and initial resolution or improvement of pneumatosis coli was achieved in 31 of 35 (89%) humans ranging in age from 10 mo to 86 yr.9,11 In a case series comparing the use of hyperbaric oxygen therapy to nonhyperbaric oxygen therapy for the treatment of pneumatosis cystoides intestinalis, patients treated with hyperbaric oxygen therapy had resolution of signs faster than those treated with nonhyperbaric oxygen therapy (mean 4.7 days versus 14.6 days, respectively).20 Because of this dog’s brief hospitalization and poor patient compliance during his second session, he only received hyperbaric oxygen therapy once. Reported duration of hyperbaric oxygen therapy for resolution of pneumatosis coli in humans is highly variable, ranging from a single session to 35 daily sessions.11 The significance of hyperbaric oxygen therapy use in the resolution of pneumatosis coli in this dog is unknown, and further prospective studies are warranted. This treatment modality was well tolerated and may be beneficial in other veterinary cases of pneumatosis coli.
Although pneumatosis coli is rarely diagnosed in dogs, this case and other published reports describe a more benign form of disease that resolves with medical management. The etiopathogenesis of this case was unknown. This dog may have been predisposed to developing pneumatosis coli as a result of long-term immunosuppressive therapy. Prognostic factors and mortality rates of dogs with pneumatosis coli have not been described, although this case indicates that the variables used to guide surgical intervention in humans may not apply in dogs.
Conclusion
Pneumatosis coli should be considered in cases with clinical signs of colorectal disease and can be diagnosed on abdominal radiographs. This condition is not well characterized in veterinary medicine and etiology is often unknown. Medical management should be considered in cases without evidence of surgical disease, such as gastrointestinal obstruction or septic peritonitis. Broad spectrum antibiotic, anthelminthic, and gastroprotectant therapies were used in this case. Hyperbaric oxygen therapy may also be beneficial in treatment of this condition in veterinary patients, and further prospective studies are warranted.
(A) Right lateral abdominal radiograph taken at presentation showing moderate-to-severe pneumatosis coli. The red arrows show gas within the colonic wall. (B) Ventrodorsal abdominal radiograph taken at presentation showing moderate-to-severe pneumatosis coli. The red arrows show gas within the colonic wall. (C) Abdominal ultrasound image showing irregular colonic wall with reverberation artifact. The red arrows show reverberation artifact due to free gas within the colonic wall.
(A) Right lateral abdominal radiograph showing improved pneumatosis coli on day 4 of hospitalization. The red arrows show improvement of free gas within the colonic wall. (B) Ventrodorsal abdominal radiograph showing improved pneumatosis coli on day 4 of hospitalization. The red arrows show improvement of free gas within the colonic wall.
Contributor Notes
