Thermoregulation in dogs is a tightly regulated process that maintains heat gain and dissipation within a narrow core body temperature range or setpoint. At Ta < 32.0°C (89.6°F), > 70% of a dog's heat loss is by radiation and convection; however, as the Ta approaches the dog's normal core body temperature, heat loss occurs primarily through evaporation by the dog panting.1,2 Additionally, this evaporative process becomes less efficient at higher RHs.1–4 If compensatory measures are insufficient and the core body temperature continues to increase above the upper reference limit, thermoregulatory breakdown eventually occurs, resulting in the clinical signs of heat exhaustion or heatstroke.3,4 Although the reference range for clinically normal Tres in dogs is 37.2°C to 39.2°C (99.0°F to 102.5°F), it is common for Tre to be > 42.0°C (107.6°F) during exercise in working and sporting breeds of dogs with no adverse effects,3,4–10 whereas core body temperatures as low as 41.0°C (105.8°F) have been associated with permanent brain damage in dogs.1–4,11 While working, Special Operations’ MWDs commonly have core body temperatures of 42.2°C (108.0°F), and handlers of MWDs evaluate the dogs’ panting efforts, rather than relying on Tre alone, for signs of thermoregulatory breakdown. For instance, handlers of MWDs assess whether a dog's panting is uncontrolled, how much the dog's lips are retracted to increase the surface area available for maximum evaporation, whether the dog has increased upper airway sounds, and whether there is any change in behavior (eg, seeking shade, reluctance to move, or both).12 Therefore, although the Tre is an important measurement in a dog that is suspected of having an HIIE, it must be interpreted in conjunction with other factors (eg, the panting assessment, BCS, type of exercise, and the dog's normal behavioral characteristics) to assess whether a measured Tre is abnormal for a particular dog at a particular instance.12–15
To conduct an accurate heat stress assessment, several exogenous (eg, environmental and activity-related) and endogenous (eg, dog-related) factors associated with heat production or dissipation should be considered. Factors such as environmental conditions (eg, high Ta and RH), type of activity (eg, exercise), BCS (eg, obesity), coat, history of a prior HIIE, level of acclimatization, brachycephalic anatomy, and laryngeal disease are associated with an increased risk for having an HIIE.2–4,13,16,17 How much each factor contributes to an HIIE in dogs is unclear because of the interactions these factors have. In people, 2 models, the HI and WBGT, have been developed and used to determine the effective temperature. The HI model uses Ta and RH, whereas the WBGT model incorporates Ta, RH, wind speed, and solar radiation. Use of these models has been suggested in dogs4; however, we were not aware of any controlled study in dogs in which these models have been applied. On the basis of physiologic processes and published experimental data in dogs, investigators have modified models used in humans to assess the effect of Ta, RH, and air movement (fan) on MWDs at rest versus active and concluded that whether at rest or active in a Ta of 35°C (95°F), an MWD could maintain a normal Tre for 130 minutes in RH of 50%, compared with 300 minutes in RH of 20%.18 With a fan on, the duration increased to 140 and 390 minutes, respectively.18
The purpose of the study presented here was to evaluate patient demographics and environmental factors for potential association with HIIEs in MWDs. We hypothesized that environmental factors, patient demographics, or both correlated with Tre and could be predictors for HIIEs in MWDs.
Materials and Methods
Animals
Veterinary treatment records from the DODMWDVS located at the Joint Base San Antonio–Lackland in San Antonio, Tex, were reviewed to identify MWDs that had an HIIE between January 2008 and December 2014 listed on their master problem list in accordance with SOP #400-13.19 Inclusion criteria were MWDs that were a GSD, LAB, or MAL and had a history of ≥ 1 HIIE but no evidence of a preexisting medical condition. For MWDs that had > 1 HIIE, data from only the first HIIE were included in analyses, except for analyses used to evaluate the odds of having > 1 HIIE; therefore, potential physiologic changes from a previous HIIE would not skew results.
Definitions
Heat-induced injury events were instances in which MWDs’ thermoregulatory processes were insufficient, resulting in abnormal clinical signs and a diagnosis of heat exhaustion or heatstroke.
Exertional hyperthermia was defined as abnormal clinical signs limited to heavy (but not dyspneic or uncontrolled) panting and a Tre usually < 41.1°C (106.0°F). In addition, MWDs could be mildly tachycardic (eg, 100 to 120 beats/min; reference range, 60 to 100 beats/min) but with a strong pulse. Exertional hyperthermia was considered within the physiologic limits of well-conditioned MWDs and was not considered a medical problem.
Heat exhaustion was defined as abnormal clinical signs of uncontrolled panting (eg, a dog that would not stop panting despite being exposed to alcohol vapors [such as on a cotton ball held before the dog's nose] or having someone blow on the dog's nose); dyspnea (especially with upper airway noise); tachycardia (generally ≥ 120 to 140 beats/min) with fair or poor pulse quality; and signs of exercise intolerance or poor stamina, alone or in combination.
Heatstroke was defined as abnormal clinical signs of heat exhaustion plus neurologic signs, including weakness, collapse, or altered mentation, alone or in combination.19
All MWD groups consisted of young, healthy dogs that were monitored and handled by trained personnel and received regular veterinary medical care. There were 4 main groups of MWDs.
Procurement MWDs were dogs in initial training and primarily procured from regions where the Ta was 11.1°C (20.0°F) cooler than the DODMWDVS site. Upon arrival, procurement MWDs were housed in outdoor kennels while awaiting enrollment in a training class. Wait times varied from 4 to 12 weeks. Exercise during this wait time consisted of 15-minute walks on a leash ≥ 3 times/wk.
Handler training aids were trained MWDs kept on-site and used in courses to train new MWD handlers.
Puppy program dogs were 10- to 18-month-old offspring from the MWD breeding program that were currently in MWD training.
Breeding program adult dogs were sexually intact male and female MWD breeding stock housed and bred at the DODMWDVS site. The breed and sex distributions of this population were unavailable, and the dogs in this group made up 2.5% of the total annual number of MWDs housed at the facility. No breeding stock were included in the analyses because the only reported data were the names of the sires and dams of dogs with HIIEs.
All MWD groups participated in the following 5 activities, except for the training activity, in which the MWD breeding program adult dogs did not participate.
Training activities involved patrol and detection skills (scouting, building and vehicle search, obedience, controlled aggression, and gunfire). All MWDs, except for adult dogs in the MWD breeding program, were required to complete 60 days of training activity.20 Training occurred Monday through Friday from 6:30 am to 11:30 am during cooler months (ie, October through May) and from 5:30 am to 10:30 am during warmer months (ie, June through September).
Spot day activities occurred once a month from 6:00 am to 10:30 am, during which all dogs were walked from their kennels to the veterinary treatment facility for measurement of body weight and administration of monthly preventatives against heart-worms, fleas, and ticks. Handlers of more aggressive dogs were required to muzzle their dogs in the dogs’ kennels, and all dogs were required to be muzzled just before the dogs entered the veterinary facility. The muzzles were either leather or leather with a hard-plastic end. Each type had multiple holes around the mouth area to allow the dog to breathe normally and had an adjustable buckle closure for proper and secure fit. To mitigate the risk of environmental heat stress on MWDs standing in long lines on spot day, all available MWD handlers would assist in bringing dogs to the veterinary facility.
Walking activities included leashed walks for exercise that consisted of 15-minute walks ≥ 3 times/wk and walks from dogs’ kennels to the veterinary facility, training area, or a trailer (well ventilated and equipped with air conditioner units) for transport to the training area.
Kennel activities included those by dogs while in their kennels, which were well ventilated, outdoor, and constructed of wire mesh on 3 sides with a covered roof and access to an enclosed area. Depending on outside activity, the dogs rested, walked, or intermittently spun or jumped while in their kennels.
Playing activities were off-leash sprint-type activities that involved retrieving toys thrown by the dogs’ handlers.
Data collection
Environmental data—The environmental data collected from the MWDs’ veterinary treatment records included Ta, RH, HI, and WBGT for the day the HIIE occurred. Data were either empirical (eg, Ta and RH) or derived from a formula (eg, HI and WBGT). All empirical environmental data were sourced from weather data available at the Joint Base San Antonio–Lackland, whereas HI and WBGTa were calculated from empirical environmental data to determine the apparent temperature (ie, how the human body perceived Ta in given humidity and sunlight conditions), a commonly considered factor relative to heat stress on the human body.4,21 The HI was calculated from a mathematical formulab that incorporated Ta and RH to determine the apparent temperature in the shade. The WBGT was deriveda from data collected with a heat stress monitorc that measured Ta (dry bulb temperature), RH, wind speed, and globe temperature (estimated mean radiant temperature resulting from visible and infrared radiation) and calculated the WBGT (as an output measurement for apparent temperature accounting for the effects of sunlight).
Related to environmental conditions, SOP #400-13 had the following restrictions for MWD handling and training.
Anytime the Ta or the wet bulb temperature was ≥ 32.2°C (90.0°F), exercise of MWDs in pool status (dogs currently not assigned to a training class), training aides, or dogs on medical hold (temporary medical condition prohibiting training) were to have ceased. Because at any given time it was possible, but not probable, for some dogs from all 4 MWD groups to be listed in pool status, the number of dogs in this status was expected to have varied through the study period.
Dogs that previously had an HIIE could be groomed or exercised only in the morning.
Training of dogs currently enrolled in an MWD training class would not be altered by a Ta or a wet bulb temperature ≥ 32.2°C; however, individual MWDs were to be assessed in their class groups and taken for veterinary evaluation when needed.19
Patient demographics—Data collected from the medical records of each MWD in the study included sex, age, breed, Tre on initial examination at the veterinary hospital, body weight, BCS (on a scale from 1 to 9), coat color, activity engaged in before examination, and number of HIIEs.
Statistical analysis
Descriptive and inferential statistics were assessed. Commercial software,d,e was used, and for all comparisons, values of P < 0.05 were considered significant. Descriptive statistics (mean ± SD) were used to summarize data in terms of the number of HIIEs by breed, coat color, sex, activity, and number of HIIEs per MWD. Simple linear regression analysis was performed and a coefficient of determination (R2) calculated to determine how well the environmental factors (Ta, RH, HI, and WBGT) and patient BCS correlated with Tre at initial examination following the HIIE. A 1-way ANOVA was performed to compare the effect of the environmental factors (Ta, RH, HI, and WBGT) on each breed. Additional ANOVA were also performed on each of the MWD demographics (sex, breed, coat color, and activity) and calendar month with patient Tre as the dependent variable. The χ2 test was used to examine the frequency distribution between MWD breed and patient demographics (Tre and activity), environmental factors (Ta, RH, HI, and WBGT), and calendar month that the HIIE occurred. Continuous data were discretized at a resolution needed to create cell frequencies of sufficient count by conventional standards (ie, 1-, 5-, and 10-unit intervals). All variables were summarized in contingency tables. Cells that contributed the most to a significant χ2 statistic, by having the largest difference between the observed and expected counts if randomized, were recorded. A Fisher exact test was performed for cells with an expected count < 5. Both Ta and RH versus the number of observed HIIEs were graphed in a bivariate histogram. For MWDs with > 1 HIIE, OR analysis was performed to evaluate the odds of having a subsequent HIIE, compared with the overall cohort of MWDs having an initial HIIE, and the χ2 test was used to examine whether breed was associated with multiple HIIEs.
Results
Animals
Of 5,584 MWDs, 103 met the inclusion criteria. Eighty of the 103 (77.7%) dogs were sexually intact males, 21 (20.4%) were spayed females, and 2 (1.9%) were castrated males. The mean ± SD age was 22.7 ± 7.5 months, and the mean ± SD body weight was 31.4 ± 4.6 kg (69.1 ± 10.1 lb). When considered by MWD groups represented, 53 of the 103 (51.5%) dogs were in initial training, 35 (34.0%) were training aides, 15 (14.6%) were MWD puppy program dogs, and none were breeding adults. By breed, there were 39 (37.9%) MALs, 32 (31.1%) GSDs, and 32 (31.1%) LABs.
HIIE classification
Four of the 103 (3.9%) MWDs had heatstroke and Tre of 42.3°C (108.1°F; n = 1), 43.1°C (109.6°F; 2), or 43.2°C (109.8°F; 1) on initial examination following their HIIE. The remaining 99 MWDs were classified as having had heat exhaustion on the basis of uncontrolled panting, which per SOP #400-13 served to distinguish heat exhaustion from exertional hyperthermia.19 The median number of dogs treated each year for an HIIE (heat exhaustion or heatstroke) at the DODMWDVS that resulted in a change in status of the MWD from fit for duty to unfit for duty during the 7-year study period was 14 MWDs (range, 5 to 29 MWDs). The duration that MWDs remained in the unfit-for-duty status from heat exhaustion was typically 3 to 4 days. Dogs with heat stroke were required to be enrolled in a 7-week heat reconditioning program before they were allowed to return to duty.19
Tres
The duration from onset of clinical signs of an HIIE to measurement of Tre at the veterinary facility was estimated to vary from 5 to 30 minutes dependent upon where the MWD was located at the time of an HIIE. Overall, the mean ± SD Tre was 41.6 ± 0.8°C (106.9 ± 1.5°F; range, 39.5°C to 43.3°C [103.1°F to 110.0°F]). The mean ± SD Tre did not differ significantly (P = 0.86) among the breeds of GSD (41.5 ± 0.8°C [106.7 ± 1.5°F]), LAB (41.6 ± 0.7°C [106.9 ± 1.3°F]), and MAL (41.6 ± 1.0°C [106.9 ± 1.8°F]; Table 1). However, when Tre was discretized at 0.6°C (1.0°F) increments and frequency distribution by breed was assessed, the proportion of dogs from each breed in each increment differed significantly (P = 0.006), with more than expected by chance alone GSDs (n = 10) represented in the 40.6°C (105.1°F) Tre increment, GSDs (11) and LABs (12) in the 41.1°C (106.0°F) Tre increment, LABs (11) and MALs (12) in the 41.7°C (107.1°F) Tre increment, and MALs (8) in the ≥ 42.8°C (109.0°F) Tre increment (Figure 1). Important to note was that 8 of 39 (20.5%) MALs were in the ≥ 42.8°C increment, the highest reported Tre group. In addition, there were fewer GSDs (n = 2) than expected in the 41.7°C Tre increment, fewer MALs (5) than expected in the 41.1°C Tre increment, and fewer LABs (4) than expected in the 40.6°C Tre increment.
Results of 1-way ANOVA used to identify potential associations of environmental factors (Ta, RH, HI, and WBGT) and Tre with breed in 103 MWDs treated for an HIIE between January 2008 and December 2014.
| GSD | LAB | MAL | |||||
|---|---|---|---|---|---|---|---|
| Variables | No. (%) of dogs | Mean ± SD | No. (%) of dogs | Mean ± SD | No. (%) of dogs | Mean ± SD | P value† |
| Ta* (n = 59) | 21 (35.6) | 27.1 ± 5.7 | 17 (28.8) | 24.6 ± 4.7 | 21 (35.6) | 26.2 ± 3.7 | 0.26 |
| RH‡ (n = 59) | 21 (35.6) | 64 ± 1.7 | 17 (28.8) | 73 ± 2.4 | 21 (35.6) | 75 ± 1.7 | 0.19 |
| HI* (n = 59) | 21 (35.6) | 29.2 ± 7.3 | 17 (28.8) | 25.9 ± 6.3 | 21 (35.6) | 27.6 ± 4.3 | 0.25 |
| WBGT* (n = 46) | 17 (37.0) | 27.9 ± 2.9 | 10 (21.7) | 27.8 ± 2.9 | 19 (41.3) | 28.6 ± 1.9 | 0.64 |
| Tre* (n = 103) | 32 (31.1) | 41.5 ± 0.8 | 32 (31.1) | 41.6 ± 0.7 | 39 (37.9) | 41.6 ± 1.0 | 0.86 |
Reported as °C.
Value of P determined for differences among the means ± SDs reported for the breeds.
Reported as percentage RH.
Histogram of the distribution of Tre discretized at 0.6°C (1.0°F) increments in the 3 breeds (GSD [yellow], LAB [green], and MAL [gray]) of 103 MWDs treated for an HIIE between January 2008 and December 2014.
Citation: Journal of the American Veterinary Medical Association 256, 7; 10.2460/javma.256.7.792
Coat color
The coat colors of the 103 MWDs were categorized as tan (n = 32), black (28), black and tan (16), sable (12), yellow (10), and brindle (5; Table 2). The mean Tre did not differ significantly (P = 0.28) among MWDs grouped according to coat color.
Results of 1-way ANOVA used to identify potential associations of MWD coat color, activity during HIIE, and month of HIIE with Tre on initial examination following HIIE in the MWDs in Table 1.
| Variable | No. (%) of MWDs | Mean ± SD Tre (°C) | P value |
|---|---|---|---|
| Coat color (n = 103) | 0.28 | ||
| Black | 28 (27.2) | 41.4 ± 0.6 | |
| Black and tan | 16 (15.5) | 42.0 ± 0.8 | |
| Brindle | 5 (4.9) | 41.8 ± 0.9 | |
| Sable | 12 (11.7) | 41.4 ± 0.7 | |
| Tan | 32 (31.1) | 41.5 ± 1.0 | |
| Yellow | 10 (9.7) | 41.9 ± 0.6 | |
| Activity (n = 102) | 0.23 | ||
| Kennel | 6 (5.9) | 41.6 ± 0.5 | |
| Playing | 3 (2.9) | 42.5 ± 0.7 | |
| Spot day | 15 (14.7) | 41.4 ± 0.5 | |
| Training | 66 (64.7) | 41.6 ± 0.9 | |
| Walking | 12 (11.8) | 41.8 ± 0.7 | |
| Month (n = 103) | 0.43 | ||
| Jan | 4 (3.9) | 41.0 ± 0.3 | |
| Feb | 3 (2.9) | 42.1 ± 1.0 | |
| Mar | 2 (1.9) | 41.8 ± 0.6 | |
| Apr | 10 (9.7) | 41.8 ± 0.7 | |
| May | 11 (10.7) | 41.8 ± 1.2 | |
| Jun | 15 (14.6) | 41.5 ± 0.8 | |
| Jul | 16 (15.5) | 41.8 ± 0.6 | |
| Aug | 17 (16.5) | 41.4 ± 0.8 | |
| Sep | 15 (14.6) | 41.5 ± 0.8 | |
| Oct | 6 (5.8) | 41.1 ± 0.7 | |
| Nov | 3 (2.9) | 41.6 ± 0.4 | |
| Dec | 1 (1.0) | 40.9 |
Sex
The sex of 101 MWDs (80 sexually intact males and 21 spayed females) was included in the analysis. Because there were only 2 castrated males in the study, they were not included in the analysis for potential association between sex and HIIE. There was no significant (P = 0.89) difference in the mean ± SD Tre between sexually intact males (41.6 ± 0.8°C) and spayed females (41.6 ± 0.9°C [106.9 ± 1.6°F]). Similarly, χ2 analysis identified no significant (P = 0.78) difference in frequency distribution of Tre between sexually intact male MWDs versus spayed female MWDs.
Activity
The activity immediately before the HIIE was available in the medical records for 102 of the MWDs. Of the 5 potential activities, 3 (training [n = 66], spot day [15], and walking [12]) accounted for 91.2% [93/102] of the HIIEs (Table 2). There was no significant (P = 0.23) difference in the mean Tre for dogs grouped according to activity at the time of the HIIE. However, when evaluated on the basis of breed and specific activities, there was a significant (P = 0.022) difference in frequency distribution of breeds for 3 of the activities. For walking, the proportion of dogs affected was higher for GSDs (7/12) versus that for LABs (2/12) or MALs (3/12). For kennel activities, the proportion of dogs affected was higher for LABs (5/6) versus that for GSDs (0/6) or MALs (1/6). At play, MALs (n = 3) were the only dogs that had HIIEs.
Time of year
When dogs were grouped according to the month during which their HIIE occurred, most (84/103 [81.6%]) HIIEs occurred in April through September, with 48 of the 103 (46.6%) HIIEs during the summer months of June through August (Table 2). However, the mean Tre of MWDs grouped by month of HIIE occurrence did not differ significantly (P = 0.43). Further, there was no significant (P = 0.39) differences among the frequency distribution of breeds represented in each month.
Linear regressions
Not all records of MWDs had the environmental data or BCS noted with each HIIE. However, data on those animals for which the factors were available were included in a simple linear regression analysis performed to determine how well each of the environmental factors (Ta [n = 59], RH [59], HI [59], and WBGT [46]; Table 1) and dog BCS (66) correlated with Tre at the time of the HIIE. No significant relationships were identified between Tre as the dependent variable and environmental factors (Ta [P = 0.42; R2 = 0.01], RH [P = 0.30; R2 = 0.02], HI [P = 0.75; R2 < 0.01], and WBGT [P = 0.11; R2 = 0.06] as the independent variables. However, linear regression analysis of BCS (mean ± SD, 5.1 ± 0.8 [on a 9-point scale]; n = 66) as the independent variable and Tre (mean ± SD, 41.6 ± 0.8°C) as the dependent variable indicated a potential relationship (P = 0.06; R2 = 0.05). Only 5 dogs had a BCS > 6.
Breed
One-way ANOVA was performed to identify potential associations between environmental factors and HIIEs in the MWDs overall and by breed. The mean ± SD Ta was 26.2 ± 4.8°C (79.2 ± 8.6°F; range, 12.2°C to 38.9°C [54.0°F to 102.0°F]; n = 59). Only 14 of 59 (24%) HIIEs occurred when the Ta was ≥ 29.4°C (85.0°F), and 29 (49%) occurred when the Ta was < 23.9°C (75.0°F). The mean ± SD Ta when MWDs had HIIEs did not differ significantly (P = 0.26) among the 3 breeds (Table 1). Further, when the Ta contingency table was discretized into 2.8°C (5°F) increments, no significant (P = 0.13) difference in the frequency distribution of affected animals on the basis of breed was identified with χ2 analysis.
The mean ± SD RH was 70 ± 19.7% (range, 22% to 100%; n = 59). A bivariate histogram plot was generated as a visual aid to show the distribution of HIIEs in relation to the environmental factors of Ta and RH (Figure 2). The number of HIIEs appeared to increase with a combination of higher Ta and RH. The highest number of HIIEs (n = 9) occurred under the combined conditions of the RH ≥ 90% and the Ta in the 21.1°C to 21.6°C (70.0°F to 70.9°F) increment, followed by 8 HIIEs when the RH was in the 60% increment and the Ta was in the 26.7°C to 27.2°C (80.0°F to 80.9°F) increment. Fifty-two of 59 (88%) HIIEs occurred at an RH > 50%. The mean RH when MWDs had HIIEs did not differ significantly (P = 0.19) among the 3 breeds (Table 1). Further, when the RH contingency table was discretized into 10% increments (eg, RH 50%, 60%, and 70%), χ2 analysis identified no significant (P = 0.57) difference in the frequency distribution of affected animals on the basis of breed.
Histogram of the distribution of 59 of the MWDs in Figure 1 by increments of Ta (°C [°F]) and by RH (%) at the time of their HIIE.
Citation: Journal of the American Veterinary Medical Association 256, 7; 10.2460/javma.256.7.792
The mean ± SD HI was 27.8 ± 6.2°C (82.0 ± 11.2°F; range, 11.1°C to 41.7°C [52.0°F to 107.0°F]; n = 59). Thirty-four of 59 (58%) HIIEs occurred when the HI was > 26.7°C. The mean HI at the time MWDs had HIIEs did not differ significantly (P = 0.25) among breeds. For the χ2 analysis, the HI contingency table was discretized into 5.6°C (10.0°F) increments, and no significant (P = 0.34) difference in the frequency distribution of affected animals was identified on the basis of breed.
The mean ± SD WBGT was 28.3°C (83.0°F; range, 20.4°C to 31.9°C [68.7°F to 89.5°F]; n = 46). Thirty-five (76%) of the HIIEs occurred at an HI > 26.7°C, whereas no HIIEs occurred at an HI < 20.0°C (68.0°F). The mean WBGT when MWDs had HIIEs did not differ significantly (P = 0.64) among breeds (Table 1). The WBGT contingency table was discretized into 2.8°C increments, and with χ2 analysis, no significant (P = 0.45) difference in the frequency distribution of affected animals was identified on the basis of breed.
MWDs with multiple HIIEs
Overall, there were 193 HIIEs in 103 MWDs during the 7-year study period. Forty-four of 103 (42.7%) MWDs had multiple HIIEs that were included in the previously mentioned analyses. Twenty-one of the 103 (20.3%) MWDs had 2 HIIEs, 12 (11.7%) had 3 HIIEs, 3 (2.9%) had 4 HIIEs, 4 (3.9%) had 5 HIIEs, and 4 (3.9%) had 6 HIIEs. The median number of HIIEs per MWD was 1 (range, 1 to 6), and the median interval to recurrence after the first HIIE was 39 days (range, 1 day to 6 years). Of the 5,584 MWDs, 103 (1.8%) sustained a first HIIE; 44 of those 103 (42.7%) MWDs sustained a second HIIE. The odds of an HIIE were significantly (OR, 39.7; 95% confidence interval, 25.7 to 61.4; P < 0.01) greater for the 103 dogs that had already had an initial HIIE (74.6%) than for dogs in the overall cohort of 5,584 MWDs to have had an initial HIIE (1.9%). There was no evidence of a significant (P = 0.69) difference in frequency distribution on the basis of breed for MWDs that had multiple HIIEs. Interestingly, an entire MAL family (sire, dam, and 5 offspring) had > 1 HIIE, with one offspring having had 4 HIIEs and another offspring having had 3 HIIEs.
Discussion
The present study focused on HIIEs in 3 breeds (MAL, GSD, and LAB) of MWDs, and factors that had been identified in other HIIE studies3,4,16,17,22 were evaluated to determine whether they increased the risk of an HIIE in these breeds. Depending on the factor being evaluated, most of the findings in the present study either were in agreement with previous findings or augmented them. For instance, our findings augmented results of studies2,7 that show when the RH is ≥ 50%, the RH appears to be a greater risk factor of an HIIE than the Ta. A notable exception in the present study was our finding that indicated greater odds of MWDs having > 1 HIIE; however, this finding was consistent with findings in people.5,21,23,24 Although investigators2,3,19 have hypothesized such a possibility, the present study was the first with supporting data. In addition, 1 family of MWDs (7 MALs [sire, dam, and 5 offspring]) was identified in which all members had > 1 HIIE during the study period. This finding supported the premise that certain genetic phenotypes may predispose dogs to HIIEs.1
A recent study20 shows that, following a training activity, the peak Tre in MWDs did not occur until 8 to 12 minutes after the activity had ended, and those investigators suggested that Tre continue to be monitored for the first 15 minutes following the activity. In the present study, exercise completion times were not recorded in the veterinary treatment records, and the duration between recognition of clinical signs of an HIIE and the dog being taken to the veterinary facility ranged from 5 to 30 minutes; therefore, the Tre measured may or may not have been the peak Tre.
Excessive body fat impairs normal heat dissipation. Obesity, and specifically a BCS > 6/9,17 is a risk factor for death in dogs with heatstroke.2 Therefore, in addition to recording the body weight of the dog, a BCS should also be provided. In the present study, results suggested a potential (P = 0.06; R2 = 0.05) linear relationship between BCS and Tre. However, only 5 MWDs had a BCS > 6, and because of their physical conditioning as MWDs, findings in these dogs may not translate to other populations of dogs.
All 5 activities were represented as the activity during which MWDs had HIIEs, and the activity of playing was associated with the highest Tre. However, only 3 MALs had HIIEs during play. An explanation for these findings was that the type of physical conditioning the MWD was undergoing was very activity specific; hence a dog performing a sprint-type work activity, versus aerobic or endurance work activities, would have a different conditioning protocol. For example, if an MAL was predominately conditioned for an endurance activity, the change to a sprint-type activity that occurred during play could have resulted in a higher core body temperature and increased the risk of an HIIE because the physiologic conditioning adaptations that had occurred for the endurance activity were only those needed to perform that conditioning activity.12,13 Rectal temperatures above the reference limit during the spot day activity were generally attributed to use of muzzles, dogs’ hyperexcitable temperaments, or environmental conditions (most notably a higher RH), alone or in combination.
In the present study, 24% of the HIIEs occurred at a Ta ≥ 29.4°C, which was not surprising because most MWD training stops at ≥ 32.2°C, per SOP.19 In addition, there was no significant difference among breeds in regard to the mean Ta when HIIEs occurred. Interestingly, in the present study, 88% of the HIIEs occurred in an RH > 50%, and 76% occurred at a WBGT > 26.7°C. More dogs had an HIIE when the Ta was > 26.7°C combined with an RH > 60%. However, no single environmental variable (Ta, RH, HI, or WBGT) was a predictor of MWD's Tre following HIIE. Because the present study did not include a distribution of weather conditions across all days, study results may not have accurately reflected a greater risk of HIIEs during certain environmental conditions, but simply indicated that these were the most common environmental conditions present when the dogs were training. In addition, multiple studies3,4,12,25 show that a lack of acclimatization may increase the risk of HIIEs in dogs. In the present study, dogs in the MWD procurement (initial training) group had the highest proportion of HIIEs, but no definitive conclusions could be drawn regarding whether a lack of acclimatization contributed to these results because the initial period of environmental acclimatization ranged from 4 to 16 weeks and because acclimatization can vary from 10 to 20 days, with some dogs requiring up to 60 days to reach full acclimatization.3,4,25,26
In the present study, there were fewer HIIEs observed during the cooler months; however, the mean ± SD Tre on initial examination following an HIIE did not differ substantially when considered across months. A study27 of the effect of seasonal changes on the heat balance in dogs acclimatized to outdoor climates shows that domestic dogs typically have narrow thermoneutral zones in both warm and cold seasons. Therefore, our finding that the mean ± SD Tre did not substantially differ throughout the year may have represented the MWDs’ normal thermoregulatory response to changes in environmental conditions.
On the basis of our findings that the odds of an MWD having a second HIIE (74.6%) were much greater than the odds of a first HIIE (1.9%) and that an entire family of 7 MWDs had > 1 HIIE, we question whether there is a possible genetic component associated with HIIEs in dogs.1 In people, it has been hypothesized (on the basis of diverse observed responses to heat-related incidences in different strains of mice and rats) that heatstroke resistance may be linked to a heritable trait that is affected by genetic and environmental factors,28 and our findings suggested that studies are warranted to investigate whether certain genetic markers predispose dogs to HIIEs. Results of such genetic studies could potentially be used to develop screening tests to identify MWDs predisposed to HIIEs and assist in the assessment of whether MWDs are fit to return to duty after having an HIIE.
Limitations in the present study included missing environmental data, unavailable information regarding duration of unfit-for-duty status for heat exhaustion and heatstroke, subjective nature of BCS scoring, lack of serum biochemical data, lack of an MWD control group, duration of acclimatization prior to training, limited data on breeding family and MWD breeding stock, 1 geographic location with no MWDs in field conditions, MWD group (procurement, training aids, MWD puppy program, and MWD breeding stock) composition variability over time, and handler variability during MWD training and in recognizing clinical signs of heat stress in dogs. Another limitation was the fact that when MWDs were reassigned to other locations, their physical medical records accompanied them, and we had limited electronic medical records for them on-site for the present study. These limitations led to various numbers of MWD data sets for variables analyzed and limited analyses of all data to the first HIIE per MWD and to 3 MWD breeds.
The present study included a RH > 50%, which allowed for assessment of heat stress at RH levels higher than in many previous studies.2,7 Results of the present study indicated that there was little difference between sexes or among breeds of MWDs (MAL, GSD, and LAB) except during play, in which the MALs were the only breed to have had HIIEs, and except that 8 of 39 (20.5%) MALs were in the ≥ 42.8°C Tre increment. No single environmental variable was identified as a significant predictor of Tre in the MWDs of the present study, and we believe that single factors alone provide too limited a view and that an index that reflected the rate of heat loss through all mechanisms (evaporative, convective, radiative, and conductive) would be most relevant. Finally, results of the present study indicated that HIIEs in MWDs were multifaceted processes involving a combination of MWD demographics and environmental factors and that controlled studies are needed to gain a better understanding of how these factors interact.
Acknowledgments
No third-party funding or support was received in connection with the present study or the writing or publication of the manuscript. The authors declare that there were no conflicts of interest.
The opinions or assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the Army or the Department of Defense. In conducting the research described in this report, the investigators adhered to the Guide for the Care and Use of Laboratory Animals as prepared by the Committee on Care and Use of Laboratory Animals of the Institute for Laboratory Animal Research, National Research Council. Any citations of commercial organizations and trade names in this report do not constitute an official Department of the Army endorsement of approval of the products or services of these organizations.
The authors thank Ms. Angela Mound for assistance with the statistical analyses and recognize all the MWDs that have died in service of our nation.
ABBREVIATIONS
| BCS | Body condition score |
| DODMWDVS | Department of Defense Military Working Dog Veterinary Service |
| GSD | German Shepherd Dog |
| HI | Heat index (National Oceanic and Atmospheric Administration human body model) |
| HIIE | Heat-induced injury event |
| LAB | Labrador Retriever |
| MAL | Belgian Malinois |
| MWD | Military working dog |
| RH | Relative humidity |
| SOP | Standard operating procedure |
| Ta | Ambient temperature |
| Tre | Rectal temperature |
| WBGT | Wet-bulb globe temperature |
Footnotes
National Weather Service. WetBulb Globe Temperature. Available at: www.weather.gov/tsa/wbgt. Accessed Jun 15, 2018.
National Weather Service Weather Prediction Center. The heat index equation. Available at: www.wpc.ncep.noaa.gov/html/heatindex_equation.shtml. Accessed Jun 15, 2018.
QUESTemp 46 Heat Stress Monitor, 3M Corp, Saint Paul, Minn.
Microsoft Excel for MAC, Analysis ToolPack, Microsoft Corp, Redmond, Wash.
SAS Enterprise Guide, version 6.1, SAS Institute Inc, Cary, NC.
References
1. Romanucci M, Salda LD. Pathophysiology and pathological findings of heatstroke in dogs. Vet Med Res Rep 2013;4:1–9.
2. Bruchim Y, Klement E, Saragusty J, et al. Heat stroke in dogs: a retrospective study of 54 cases (1999–2004) and analysis of risk factors for death. J Vet Intern Med 2006;20:38–46.
3. Flournoy WS, Wohl JS, Macintire DK. Heatstroke in dogs: pathophysiology and predisposing factors. Compend Contin Educ Pract Vet 2003;25:410–418.
4. Johnson SJ, McMichael M, White G. Heatstroke in small animal medicine: a clinical practical review. J Vet Emerg Crit Care (San Antonio) 2006;16:112–119.
5. Ilkiw JE, Davis PE, Church DB. Hematologic, biochemical, blood-gas, and acid-base values in Greyhounds before and after exercise. Am J Vet Res 1989;50:583–586.
6. Matwichuk CL, Taylor SM, Shmon CL, et al. Changes in rectal temperature and hematologic, biochemical, blood gas, and acid-base values in healthy Labrador Retrievers before and after strenuous exercise. Am J Vet Res 1999;60:88–92.
7. Robbins PJ, Ramos MT, Zanghi BM, et al. Environmental and physiological factors associated with stamina in dogs exercising in high ambient temperatures. Front Vet Sci 2017;4:144.
8. Steiss J, Ahmad HA, Cooper P, et al. Physiologic responses in healthy Labrador Retrievers during field trial training and competition. J Vet Intern Med 2004;18:147–151.
9. Steiss JE, Wright JC. Respiratory alkalosis and primary hypocapnia in Labrador Retrievers participating in field trials in high-ambient-temperature conditions. Am J Vet Res 2008;69:1262–1267.
10. Taylor CR, Schmidt-Nielsen K, Dmi'el R, et al. Effect of hyperthermia on heat balance during running in the African hunting dog. Am J Physiol 1971;220:823–827.
11. Oglesbee MJ, Alldinger S, Vasconcelos D, et al. Intrinsic thermal resistance of the canine brain. Neuroscience 2002;113:55–64.
12. Baker J, Miller L. The effects of environmental extremes on working dogs: a collaborative initiative. US Army Med Dep J 2013;(Jan–Mar):22–27.
13. Davis M. Canine physiological resilience IED detection dog 2.0. Stillwater, Okla: Center for Veterinary Health Sciences, Oklahoma State University, 2013.
14. Angle TC, Gillette RL. Telemetric measurement of body core temperature in exercising unconditioned Labrador Retrievers. Can J Vet Res 2011;75:157–159.
15. Dampney RAL, Hales JRS. Relationship between blood pressure alterations and bursts of panting in the conscious dog: a comparison of the effects of exercise and heat exposure. J Physiol 1974;241:423–436.
16. Leon LR, Bouchama A. Heat stroke. Compr Physiol 2015;5:611–647.
17. Segev G, Aroch I, Savoray M, et al. A novel severity scoring system for dogs with heatstroke. J Vet Emerg Crit Care (San Antonio) 2015;25:240–247.
18. Berglund LG, Yokota M, Santee WR, et al. Predicted thermal responses of military working dog (MWD) to chemical, biological, radiological, nuclear (CBRN) protective kennel enclosure. Technical Report T11-03. Natick, Mass: US Army Research Institute of Environmental Medicine, 2011.
19. DODMWDVS. Military working dog SOP #400–13. Washington, DC: Department of the Army, 2017;1–7.
20. O'Brien C, Karis AJ, Tharion WJ, et al. Core temperature responses of military working dogs during training activities and exercise walks. US Army Med Dep J 2017;(3–17):71–78.
21. Sawka MN, Leon LR, Montain SJ, et al. Integrated physiological mechanisms of exercise performance, adaptation, and maladaptation to heat stress. Compr Physiol 2011;1:1883–1928.
22. Drobatz KJ, Macintire DK. Heat-induced illness in dogs: 42 cases (1976–1993). J Am Vet Med Assoc 1996;209:1894–1899.
23. Leon LR. Pathophysiology of heat stroke. In: Granger DN, Granger JP, eds. Colloquium series on integrated systems physiology. Oxford, England: Morgan & Claypool Life Sciences, 2015;1–99.
24. Kodesh E, Horowitz M. Soleus adaptation to combined exercise and heat acclimation: physiogenomic aspects. Med Sci Sports Exerc 2010;42:943–952.
25. Moseley PL. Heat shock proteins and heat adaptation of the whole organism. J Appl Physiol 1997;83:1413–1417.
26. Bruchim Y, Aroch I, Eliav A, et al. Two years of combined high-intensity physical training and heat acclimatization affect lymphocyte and serum HSP70 in purebred military working dogs. J Appl Physiol 2014;117:112–118.
27. Sugano Y. Seasonal changes in heat balance of dogs acclimatized to outdoor climate. Jpn J Physiol 1981;31:465–475.
28. Furuyama F. Strain difference in thermoregulation of rats surviving extreme heat. J Appl Physiol 1982;52:410–415.
