Evaluation of Gastroesophageal Reflux in Anesthetized Dogs with Brachycephalic Syndrome
ABSTRACT
Brachycephalic airway syndrome may predispose to gastroesophageal reflux (GER) because of the high negative intrathoracic pressures required to overcome conformational partial upper airway obstruction. To investigate this, 20 dogs presenting for elective correction of brachycephalic airway syndrome (cases) and 20 non-brachycephalic dogs (controls) undergoing other elective surgeries were prospectively enrolled. Dogs underwent a standardized anesthetic protocol, and esophageal pH was monitored. Signalment, body weight, historical gastrointestinal and respiratory disease, complete blood count, serum biochemical values, radiographic findings, and anesthetic and surgical time were compared between cases and controls, and dogs that did and did not have basic (pH > 7.5), acidic (pH < 4), or any GER. Controls had higher mean esophageal pH (6.3) compared to cases (5.6, P = .019), but there was no difference in % with GER (cases 60%, controls 40%, P = .34). When all dogs were evaluated, dogs with GER had increased creatinine (P = .01), % positive for esophageal fluid on radiographs (P = .05), and body weight (P = .04) compared to those without GER. GER was common in both cases and controls, and cases had lower esophageal pH; however, greater numbers are required to determine if a true difference exists in % GER.
Introduction
Functional partial upper airway obstruction is a common complaint in brachycephalic dogs that results from a syndrome of abnormalities including stenotic nares, hypoplastic trachea, elongated soft palate and secondary changes to include redundant oropharyngeal soft tissue, everted laryngeal saccules, and laryngeal collapse. Commonly, brachycephalic dogs presenting with respiratory signs are documented to have concomitant gastrointestinal (GI) abnormalities including ptyalism, regurgitation, and vomiting.1,2 This association was documented in one study in which 97.3% of brachycephalic dogs with respiratory signs had esophageal, gastric, or duodenal abnormalities present during endoscopic examination.1 Brachycephalic dogs are also known to be predisposed to hiatal hernia and subsequent gastroesophageal reflux (GER).3,4 Finally, in an experimental model of induced partial upper airway obstruction, GER was related to increased negative intrathoracic pressure and increased inspiratory effort.5
Reflux esophagitis, which can be caused or exacerbated by perioperative GER, may lead to consequences such as postoperative discomfort, continued or increased regurgitation, or esophageal stricture.6 This is important as a 23% mortality rate was reported in dogs with post-anesthetic esophageal dysfunction.7 Recurrent GER has also been documented as a cause of laryngeal disorders in humans8 and, recently, as an etiology for laryngeal dysfunction in a dog.9 In addition, GER severe enough to result in regurgitation (gastric contents refluxing to the oropharynx) may result in aspiration and secondary pneumonia. Pneumonia may be devastating in brachycephalic breeds, which commonly have hypoplastic tracheae and decreased airway clearance mechanisms and may be at greater risk for respiratory fatigue given their conformational abnormalities.10–12 With this in mind, understanding the frequency and character of GER in all dogs, especially those with brachycephalic airway syndrome (BAS), is critical.
Studies using an esophageal pH monitoring system to evaluate perioperative esophageal pH allow for real-time, continuous display of esophageal pH. A pH monitoring system has been used in several veterinary clinical studies evaluating GER, which mainly evaluate various anesthetic regimens on GER.13–15 Wilson et al. used a perioperative pH monitoring system to evaluate the effect of preoperative morphine,13 meperidine,14 and anesthetic gases on the occurrence of GER.15 The authors reported that the use of morphine, but not the occurrence of vomiting associated with morphine, increased the risk of GER; meperidine reduced GER; and all of the commonly used anesthetic gases had similar negligible effects on incidence of GER.13–15
There has never been objective documentation of an increased occurrence of GER during anesthesia in dogs with BAS compared to dogs without BAS. The goals of the study presented here were to document the occurrence of GER in anesthetized brachycephalic dogs, to compare this to a control population of non-brachycephalic dogs without respiratory signs, and to identify preoperative factors, such as those associated with signalment or preoperative biochemical parameters, that may alert clinicians to increase risk of GER in particular patients. Our hypothesis was that the percentage of brachycephalic dogs with GER would be greater than control non-brachycephalic dogs without conformational partial upper airway obstruction.
Materials and Methods
Animals
Brachycephalic dogs presenting for corrective surgery of partial upper airway obstruction resulting from the components of brachycephalic syndrome (cases) were prospectively enrolled in the study. Partial upper airway obstruction was confirmed by an anesthetized upper airway examination. Control dogs were client-owned, non-brachycephalic dogs that presented to the clinic for a variety of surgical procedures and did not have partial upper airway obstruction or a history of partial upper airway obstruction, based on historical absence of clinical signs associated with upper airway obstruction and a physical examination. Dogs were excluded as either a case or control animal if the thorax or abdomen was opened during surgery, if there were historical or current mast cell tumor(s), or if the dog was currently receiving a corticosteroid, non-steroidal anti-inflammatory, prokinetic, or gastric pH modifying drug. Cases or controls were also excluded if there was evidence of substantial organ dysfunction based on the following parameters: creatinine > 2.5 mg/dL, HCT < 30%, TP < 4 g/dL, bilirubin > 0.5 mg/dL, RBC < 4 × 106/μL, albumin < 2 g/dL, calcium > 12 mg/dL, glucose > 200 mg/dL, alkaline phosphatase > 500 U/L). The signalment of all dogs in the study was recorded.
Preoperative Staging
Owners completed a questionnaire upon each dog's enrollment into the study regarding the pet's history of respiratory and gastrointestinal signs. These data were used to assign a GI and respiratory grade modified from Poncet et al. based on the frequency of owner-reported clinical signs.1 The GI signs owners were questioned about included vomiting, retching, regurgitation, ptyalism, constipation, eructation, and flatulence. The respiratory signs owners were questioned about included snoring, inspiratory effort, exercise intolerance, and syncope. If these signs did not occur, the dog received a 0; if any of the above clinical signs occurred less than once per month, the dog received a 1; if any occurred more than once a month but not daily, the dog received a 2; and if any occurred daily, the dog received a 3. Respiratory and GI data were also evaluated on a binary scale based on a yes or no question of whether the dog had any history of respiratory or gastrointestinal clinical signs.
Brachycephalic dogs were assigned a brachycephalic grade, with 1 point assigned for each of the following: stenotic nares, hypoplastic trachea, elongated soft palate, everted laryngeal saccules, and grade II or III laryngeal collapse (maximum score = 5). Other data recorded included age, weight, body condition score, duration of anesthesia and surgery, preoperative hematocrit, white and red blood cell count, serum total protein, albumin, glucose, and creatinine concentrations. The attending veterinarian on each case assessed body condition score on a scale of 1–9 or 1–5. To standardize the body condition score methods, compared data were the quotient of either system to result in a body condition score of 0–1, so a body condition score of 4/5 would result in a score of 0.8.
Two-view thoracic radiographs in brachycephalic and control dogs were reviewed by a single radiologist blinded to patient data. Radiographs were evaluated for the presence of interstitial or alveolar patterns, increase in esophageal conspicuity and size (interpreted as esophageal fluid), esophageal gas, gastric gas, hiatal herniation, vertebral malformations, and the trachea diameter to thoracic inlet diameter ratio. Radiographs were taken during the same hospitalization period, generally the day prior to anesthesia and surgery. The timing in relation to anesthesia and sedation used to obtain radiographs were not standardized. Severity of radiographic signs for each finding evaluated were assigned a semi-quantitative score (0–3; 0 = unaffected, 1 = mild changes, 2 = moderate changes, 3 = severe changes).
Anesthesia and Surgery
All dogs in the study underwent a standardized anesthetic protocol of midazolam (0.1 mg/kg), hydromorphone (0.2 mg/kg), and atropine (0.02 mg/kg) IM as premedication, induction with propofola (5–7 mg/kg IV to effect following intravenous catheterization), and maintenance of anesthesia with isofluraneb delivered in 100% oxygen following intubation. Airway examination was performed immediately prior to intubation and surgical correction in all brachycephalic dogs. Surgery for correction of brachycephalic airway abnormalities was performed by one of several board-certified surgeons or residents in a surgery training program using standard techniques (either a carbon dioxide laser or sharp excision and mucosal apposition); laryngeal saccules were excised without closure.16 Several cases and controls had multiple procedures performed under anesthesia; this was recorded.
pH Measurement
Anesthetic GER was documented by continuous monitoring of esophageal pH with a pH monitorc connected to a single-channel, single-use pH probed orad to the lower esophageal sphincter. Per manufacturer's instructions, the probe was calibrated in pH 7 and pH 4 buffer solutionse immediately prior to use. The probe was placed by a single surgeon after induction of anesthesia. Placement of the probe was performed as previously described by measuring the distance from the incisors to the cranial margin of the tenth rib; the pH probe was inserted orally and advanced this distance to result in a predictable location just proximal to the lower esophageal sphincter.17 After confirmation of probe position with a lateral thoracic radiograph, white medical tape was fixed to the probe at the level of the first premolar and stapled to the dog's upper lip using surgical staples. Esophageal pH was continuously monitored and recorded at 5-min intervals to allow for a comparison of general trends; if a change consistent with GER occurred, the time was noted separately. Using criteria consistent with previous studies, prolonged (>20 sec) decreases (<4.0) or increases (>7.5) in pH were considered a change consistent with acidic or basic GER, respectively.13–15 Time to first GER was calculated as the time from the first recorded pH value to the first value consistent with GER. The pH probe was removed just prior to patient endotracheal extubation and subsequent recovery from anesthesia.
Statistics
Data were organized into comparisons of cases versus controls with acidic, basic, or any type of GER. Additionally, comparisons of data were made between dogs with or without acid, basic, or any GER regardless of case or control status. Continuous data (age, weight, red blood cell concentration, hematocrit, white blood cell concentration, serum total protein, serum albumin concentration, serum creatinine concentration, serum glucose concentration, total anesthesia time, total surgical time, body condition score, respiratory grade, GI grade, brachycephalic grade, radiographic scoring) were analyzed for normality with a Kolmogorov-Smirnov test. Data were then compared between groups with an unpaired Student's t test or a Mann-Whitney U test for parametric or nonparametric data, respectively. Data expressed as percentages (owner-reported history of breathing or GI problems, whether dog was an English bulldog, whether additional surgery was performed at the time of the surgery to address the upper airway surgery) were analyzed using a Fisher's exact test.
A repeated measures model that recognized multiple observations as belonging to the same animal was used to test for differences in pH between cases and controls and differences over time. The full model included factors for group, a continuous factor of minutes (with separate slopes for each group), and a random factor of animal. An unstructured covariance structure was used in all repeated measures models. Additionally, a homogeneity-of-slopes repeated measures regression was used to test for an interaction of minutes and group and test the assumption of using the same slope for each dog.
All data analyses were performed using statistical software.f,g A post hoc power analysis was performed using an online power calculator.h All hypothesis tests were two-sided, and the significance level was α = 0.05.
Results
Animals
Twenty brachycephalic dogs undergoing surgery for partial upper airway obstruction were enrolled as cases. Breeds included English bulldog (n = 15), French bulldog (n = 2), pug (n = 2), and Boston terrier (n = 1). Twenty control dogs were composed of 15 different breeds; these included dachshund (n = 3), mixed breed (n = 3), golden retriever (n = 2), and one each of the following: giant schnauzer, cocker spaniel, Labrador retriever, beagle, bichon frise, dachshund, Jack Russell terrier, goldendoodle, Pit bull terrier, springer spaniel, Irish wolfhound, Maltese, corgi, and a miniature pinscher. Cases were significantly younger than control dogs (Table 1, median age of cases was 2.75 yr [range 0.75–7], median age of controls was 8.5 [range 1–15 yr]; P < .0001). No significant differences were detected in weight, body condition score, red blood cell concentration, hematocrit, white blood cell concentration, serum total protein, serum albumin concentration, serum creatinine concentration, serum glucose concentration, total anesthesia time, or total surgical time between cases and controls (Table 1).
Body condition scores were compared as quotients of values assigned either on a 1–5 or 1–9 scale.
Preoperative Staging
The median brachycephalic grade for cases was 3 (range 2–3). Owner-reported respiratory signs occurred with relatively high frequency in cases (18/20 dogs), resulting in a median respiratory grade of 3 (range 0–3). There was no significant difference in the owner-reported history of GI signs between cases and controls (9/20 cases; 5/20 controls); however, cases had a significantly worse owner-reported GI grade (median GI grade in cases = 0 [range = 0–3], GI grade in control dogs = 0 [range = 0–2]; P = .004, Figure 1) compared to control dogs, indicating that GI signs occurred with greater frequency in this population. GI signs in controls included diarrhea (n = 3), retching (n = 1), and vomiting (n = 1); no dog had more than one GI clinical sign, and 4/5 dogs were only affected every few months. Among cases, GI signs included regurgitation (n = 6), vomiting (n = 4), eructation (n = 3), flatulence (n = 3), diarrhea (n = 2), retching (n =1), and constipation (n = 1). Of the nine cases with GI signs, four had more than one clinical sign, and six were affected several times a week or more.
Citation: Journal of the American Animal Hospital Association 53, 1; 10.5326/JAAHA-MS-6281
Anesthesia and Surgery
No differences were detected in anesthesia and surgical time between cases and controls (Table 1). The most common surgical procedures in cases were staphylectomy (n = 20), laryngeal saccule excision (n = 20), and correction of stenotic nares (n = 18). Sialadenectomy (n = 1) and tonsillectomy (n = 1) were also performed. The following surgical procedures were performed on controls (2 dogs had >1 procedure): skin/subcutaneous mass excision (n = 5), anal sacculectomy (n = 5), orchiectomy (n = 3), total ear canal ablation (n = 3), wound exploration (n = 2), and one case each of sialadenectomy, mandibulectomy, thyroidectomy, and perineal herniorrhaphy.
Presence of GER in Cases and Controls
There were no significant differences between control (mean change = −0.4) and case (mean change =−0.2) when change from baseline (P = .8294) pH values were assessed. The mean esophageal pH at baseline for the cases was 5.7 ± 0.34 and for the controls was 6.7 ± 0.18. Control pH values (least square mean = 6.3) were significantly higher than case pH values (least square mean = 5.6) (P = .0190, Figure 2).
Citation: Journal of the American Animal Hospital Association 53, 1; 10.5326/JAAHA-MS-6281
There was a significant interaction of minutes and group for the values (P < .0001), which supports the use of the separate slopes model. There was not a significant interaction of minutes and group for the change from baseline values (P = .1785). However, because the P-value was less than 0.20, a separate slopes model was used. For a parallel slopes model (same slope assumed for all dogs), the case and control groups were still not significantly different (P = .3654).
Anesthetic GER was present in 20/40 (50%) of all dogs, 12/20 (60%) of cases, and 8/20 (40%) of control dogs. Acidic GER was found in 12/40 dogs (8 brachycephalic, 4 control), basic GER in 6/40 dogs (3 brachycephalic, 3 control), and both acidic and basic GER in 2 dogs (1 brachycephalic, 1 control). There was no significant difference between percentage of brachycephalic and control dogs that had acid, basic, or any (either acid or basic) GER (P = .32, 1.0, .34, respectively). Based on these results, a post hoc power analysis was performed that demonstrated a sample size of 97 animals would be required to demonstrate a difference between cases and controls with any GER, and 89 animals would be required to demonstrate a difference with acidic GER, assuming a statistical power of 0.8 and α = 0.05. Thus, the study was underpowered to detect a statistically significant difference of the magnitude we demonstrated.
Duration of any GER (pH <4 or >7.5) in brachycephalic dogs was not significantly different between cases or controls (median GER duration in cases = 47.5 min [range 3–195], GER duration in controls = 32.5 min [range 5–150]; P = .18). Time to GER was not significantly different between cases and controls (cases median time to GER = 10 min [range 5–125]; control = 12.5 min [range 5–125]; all dogs = 10 min [range 5–125], P (case versus controls) = .6067).
Brachycephalic Dogs with and without GER
When data were compared between brachycephalic dogs with and without any GER, a higher radiographic score for esophageal gas was found in brachycephalic dogs without GER (median esophageal gas score = 2 [range 0–3] with GER; P = .03) compared to those with GER (median esophageal gas score = 1 [range 0–2]). Brachycephalic dogs with basic GER had a significantly higher radiographic score for an interstitial pulmonary pattern (median interstitial pattern score = 3 [range 1–3]) compared to dogs without basic GER (interstitial pattern score = 0.5 [range 0–2]; P = .01). No significant differences in any item evaluated were detected between brachycephalic dogs with acid GER compared to those without acidic GER.
Control Dogs with and without GER
Control dogs with acidic GER did not have any significant differences in variables recorded compared to control dogs without acid GER. Control dogs with basic GER had significantly higher serum creatinine concentrations (serum creatinine concentration = 1.8 ± 0.99 mg/dL) compared to dogs without basic GER (creatinine = 0.83 ± 0.23 mg/dL; P = .002). Dogs with basic GER also had a significantly higher radiographic esophageal fluid score (median esophageal fluid score = 1 [range 1–2]) compared to dogs without basic GER (esophageal fluid score = 0 [range 0–1]; P = .002). A higher serum creatinine concentration and radiographic esophageal fluid score were also present when control dogs that had any GER were compared to control dogs without any GER (P = .01 and P = .04, respectively; see Table 2).
Unless otherwise noted, the number in the box represents the P-value. Bold values are not significant.
GER, gastroesophageal reflux.
All Dogs with and without GER
Significant differences that were detected between all dogs with and without GER are listed in Table 2. Serum creatinine concentration, radiographic esophageal fluid score, and body weight were higher when dogs with GER were compared to those without GER.
Discussion
In the dogs reported here, brachycephalic dogs had a reduced overall esophageal pH compared to control dogs. GER occurred in 50% of dogs but was not shown to be more common in those dogs with BAS. Based on a post hoc power analysis, approximately 5x the number of animals would be required to demonstrate a difference in % GER between cases and controls. Clinically useful parameters to predict GER were not identified.
Previously reported rates for GER in the literature have varied widely from 16% to 60%.13,18 In a study that used healthy dogs undergoing elective surgery premedicated with varying doses of morphine, the percentage of dogs that had GER was 50% and 60% for doses of 0.22 mg/kg and 1.1 mg/kg morphine, respectively, which is similar to the occurrence reported for our control (40%) and case (60%) populations.13
The median time to first GER event was 10 min after induction of anesthesia, suggesting drugs used for premedication may have played a significant role in GER. Importantly, this also highlights a potential area of therapeutic or preventative intervention and dovetails with Wilson's previous findings that preoperative drugs have a significant influence on GER, while various anesthetic gases do not.13–15 This is further substantiated by our findings that esophageal pH data did not change significantly over time. These previous studies also found GER to occur relatively early during an anesthetic event; in one study, the median time to first GER after induction was also 10 min,13 and in another study, the mean time to first GER after probe placement was 36 min.15 The standardized anesthetic regimen employed in these dogs called for a relatively high dose of hydromorphone (0.2 mg/kg) as a premedicant, which may have increased the number of dogs with anesthetic GER. A lower dose of hydromorphone may be indicated in patients believed to be predisposed to GER.
Based on the results of this study, historical and clinical data were not useful in identifying brachycephalic dogs predisposed to GER. Although increased gas within the esophagus on radiographs was more common in brachycephalic dogs without GER, there was overlap in patient scores for esophageal gas between dogs with and without GER, making the usefulness of esophageal gas as a predictive tool low. Additionally, preoperative radiographs were not standardized with respect to patient sedation; it is possible that the use of a particular sedation protocol, or the lack of sedation and subsequent stress, may have led to aerophagia and gas within the esophagus.
Esophageal fluid seen on radiographs was more common in all dogs that had GER. This radiographic sign may indicate that GER is taking place in an awake or sedated animal. Similarly, fluid within the esophagus may be due to underlying esophagitis that is consistent with an animal that has chronic GER and is also likely to have this problem while anesthetized. However, it is difficult to ascertain whether increased visibility of the intrathoracic esophagus is really due to fluid accumulation. The increased soft tissue opacity, conspicuity, and size in the caudal esophagus was the determining radiographic sign influencing the esophageal fluid grade, but this increased opacity may also relate to esophageal wall thickness, whether the esophagus is flattened in the plane of the radiograph, or whether decreased surrounding connective tissue also played a part in some of the dog's increased esophageal prominence. The presence of esophageal fluid was not confirmed with esophagoscopy, which may result in an overstatement of the presence of fluid.
Weight and creatinine were the other two variables identified as significantly higher among all dogs with GER. A previous study evaluated the effect of weight on the occurrence of GER, and, in contrast to this study, weight was not found to be associated with a higher risk of GER in a population of otherwise healthy dogs.13 In humans, obesity is well established as being correlated with GER disease symptoms.19 In our study, however, body condition score was not associated with GER, only greater body weight that was in many cases associated with an appropriate body condition score for the given breed of dog.
The serum creatinine concentrations associated with dogs that had GER were within the laboratory reference interval, with one exception; this dog had a history of chronic kidney disease and had a mildly elevated creatinine of 2.5 mg/dL. Azotemia secondary to renal insufficiency may have played a role in GER in this case. However, given that the remainder of study subjects had creatinine values within the reference range, it is not possible to extrapolate regarding kidney disease and anesthetic GER. In people with chronic renal failure, the prevalence of GER disease is higher than healthy adults; in one study, 34% of adults with CRF had GER compared with a reported prevalence of GER in a healthy population of approximately 16%. In addition, prevalence of GER increased in symptomatic cases (44%) and if the patient underwent hemodialysis (50%).20 Given the relatively small difference in serum creatinine concentration between dogs with GER compared to those without, as well as the lack of dogs with abnormal creatinine concentrations, more data is necessary prior to making definitive clinical conclusions regarding the relationship of creatinine to GER in dogs.
A significant difference in age between control and brachycephalic dogs existed in this study. Increased age has previously been correlated with increased occurrence of GER; however, in our data as well as in a study by Wilson et al., age was not shown to be significantly associated with the occurrence of GER.13,21 As age was not associated with GER in either group, the difference in age between groups is likely of limited importance in evaluating the remainder of the data.
Patient positioning was not similar between cases and controls. A previous study found that body position in dogs (lateral, sternal, dorsal) as well as the tilt of the head up or down did not affect the occurrence of GER.21 That study also found that intra-abdominal procedures were associated with more GER events; due to this, animals undergoing intracavitary (thoracic, abdominal) surgeries were excluded from participation in the current study.21 Additionally, because of the exclusion of dogs that received corticosteroids prior to the anesthetic episode, dogs that presented in respiratory distress were likely excluded from the study, thus likely biasing the population presented here towards a less severely affected cohort. Inclusion of brachycephalic dogs without clinically significant BAS may have been an alternate control population. We elected to use non-brachycephalic dogs without partial airway obstructions so as to maximize our ability to detect differences and remove confounding influences of brachycephalic dogs with minimal or clinically undetected abnormal airway resistance. This study did not evaluate perioperative outcomes following the anesthetic period for animals that had GER. Therefore, the true clinical significance of GER in this population remains unknown and special anesthetic protocols may still be recommended, especially in light of the lower baseline esophageal pH in these dogs.
The probe used for esophageal monitoring in this study is similar to those used in the majority of veterinary research studies, in that it measures changes in pH over time. Another study, however, described the use of impedance monitoring in addition to pH monitoring to capture GER events associated with subtle changes in pH. In that study, which evaluated healthy dogs treated with placebo, cisapride, or esomeprazole, 43.3% of GER events were weakly acidic (4.0 < pH < 7.0).22 Although no dogs in the study reported here were treated with pH modifying drugs in the preoperative period, it may be that episodes of anesthetic GER unassociated with a significant change in pH went undetected, resulting in an artificially low number of GER events.
Conclusion
This study documents that GER occurs commonly in anesthetized brachycephalic and non-brachycephalic dogs. Reliable clinically significant predictors of GER were not identified in these cohorts. We were unable to prove our hypothesis that a higher percentage of brachycephalic dogs would have GER. However, brachycephalic dogs did have a reduced esophageal pH. Future studies with increased numbers of cases and controls are indicated.
Scatter plot of owner-reported GI grades in brachycephalic dogs (n = 20, cases) compared to non-brachycephalic dogs (n = 20, controls). The GI signs owners were questioned about included vomiting, retching, regurgitation, ptyalism, constipation, eructation, and flatulence. If these signs did not occur, the dog received a 0; if any of the above clinical signs occurred less than once per month, the dog received a 1; if any occurred more than once a month but not daily, the dog received a 2; and if any occurred daily, the dog received a 3. Cases had a significantly worse owner-reported GI grade (*, P = .004). The bar represents range; the median value for both groups is 0.
Mean ± standard error of the mean esophageal pH in brachycephalic dogs (cases) undergoing general anesthesia and surgery for correction of upper airway obstruction due to brachycephalic syndrome (n = 20, grey line) or non-brachycephalic control dogs undergoing general anesthesia and surgery for other reasons (n = 20, black line). Control pH values (least square mean = 6.3) were significantly higher than case pH values (least square mean = 5.6) (P = .0190). There were no significant differences between control (mean change = −0.4) and case (mean change = −0.2) when change from baseline (P = .8294) pH values were assessed.
Contributor Notes
