Heatstroke and brachycephalic dogs – is there an increased risk?

[...]

Appraisal, application and reflection

Heat-related illness (HRI) and heatstroke are serious and potentially fatal conditions in dogs (Bruchim et al., 2017).  The influence of breed and skull morphology are often discussed as risk factors for the development of HRI (Hall et al., 2020) and dogs with brachycephaly are often thought to be less able to cope with heat stress (Davis et al., 2017). This is supported by evidence appraised in this summary, which suggests that brachycephalic dogs are at increased risk of HRI compared to non-brachycephalic dogs.

The evidence provided by the summarised papers is moderate. Although there are no systemic reviews currently available for analysis, four studies were identified of relevance to the question: two retrospective cohort studies, one prospective cohort study, and one case-control study. Hall et al. (2020) has a large sample size and is likely to be representative for veterinary professionals in the UK as the study utilised clinical records from primary care practice. The study focused specifically on risk factors and development of HRI and identified skull shape as a significant factor for increased odds of exertional, environmental, and vehicular HRI, reporting that this was more significant than breed classification. However, other factors were also found to influence risk, including age and body weight. This was in line with the findings of Davis et al. (2017) which reported body condition score had a greater influence on response to heat stress than breed. It is important for this to be noted when making any recommendations with regards to breed risk for HRI, as the condition is often multifactorial (Hall et al., 2021).

Segev et al. (2015a) and Segev et al. (2015b) used alternative outcome measures, but they had smaller sample sizes and are less generalisable, recording findings from one referral population, and their findings include high numbers of brachycephalic dogs affected by HRI (36/126 [29%] of dogs in Segev et al. 2015a and 10/30 [33%] in Segev et al. 2015b). While Segev et al. (2015b) does not provide a comparison of prevalence of brachycephalic breeds within the wider population, the high percentage of dogs that were classified as brachycephalic in the study does suggest that they were commonly presented with HRI, and this is strengthened when contextualised alongside the findings of Segev et al. (2015a) and Hall et al. (2020) which contain comparisons showing an increased risk. Segev et al. (2015a) has stronger evidence for the increased risk in brachycephalic dogs as it provides a comparison of the percentage of brachycephalic dogs presenting with HRI, compared to the percentage presenting for other conditions at the hospital, making these results more generalisable to the wider population and reducing the chance of potential bias due to breed popularity.

In addition, Segev et al. (2015a) and Segev et al. (2015b) selected patients that presented with more severe HRI. The data has been collected from an emergency referral hospital which will be less likely to see mild cases of HRI compared to primary care practice, and the inclusion criteria selected for patients with severe clinical signs of HRI. The fact that brachycephlic dogs made up a high percentage of cases, and were shown to be over-represented compared to the general population in Segev et al. (2015a), points to brachycephalic dogs being at higher risk of clinically significant or severe HRI, although further evidence for direct comparison of severity of HRI in brachycephalic vs non-brachycephalic breeds would be needed to confirm this impact.

The findings from Davis et al. (2017) indicated that brachycephalic dogs were more sensitive to heat stress and experience a more profound thermoregulatory response, such as higher respiratory rates or development of respiratory distress, compared to non-brachycephalic dogs.

However, there are some limitations to interpretation of these studies.  It should be commented that there is a lack of standardised definitions for the condition and clinical signs of HRI, both between the papers and within the veterinary field. Many terms, including heatstroke, HRI and heat stress are used to describe the condition and the clinical signs of HRI can be non-specific, especially in the early stages. While scoring systems have been proposed (Segev et al., 2015a; and Hall et al., 2021) these have yet to be more widely utilised or validated. This weakens the strength of evidence and may lead to selection bias; especially as certain breeds may show more overt clinical signs related to HRI without the underlying pathology of significant non-pyrogenic hyperthermia. Further study and a validated scoring system for HRI events would enable stronger evidence for risk factors (including brachycephaly) for HRI to be generated.

The impact of the lack of a universal clinical definition or scoring for HRI is further compounded when looking at brachycephaly. Brachycephalic breeds may more frequently demonstrate the early clinical signs of HRI, such as panting, especially those that are suffering from Brachycephalic Obstructive Airway Syndrome (BOAS) (Packer et al., 2015), as they may be less tolerant of exercise (Lilja-Maula et al., 2017). This may increase the likelihood of a HRI diagnosis, based on their clinical signs alone without a meaningful increase in body temperature. Although Davis et al. (2017) attempted to address this with their ‘cool’ environment and found that in this environment there was no significant difference in the resting respiratory rates of brachycephalic and non-brachycephalic dogs, this study did not account for dogs that had been diagnosed with BOAS. Moreover, none of the other studies included BOAS as an independent factor for development of HRI, and several studies excluded dogs with concurrent conditions which may have included BOAS, so it is impossible to determine whether the effects of BOAS may have led to selection or interpretation bias. Further study to investigate the relationship of BOAS as a specific risk factor for HRI, including looking at the relationship between BOAS classification and severity of HRI, would help to determine if brachycephaly alone is a risk factor for HRI or whether BOAS is more relevant.

Some of the impact from the lack of definition for HRI can be mitigated when considering the inclusion and exclusion criteria in the studies. Hall et al. (2020) used a broad definition of heatstroke and included all recorded presentations classified as HRI, from mild to severe clinical signs. This means it is likely that the breadth of the different definitions for HRI were likely captured within these broad inclusion criteria. In addition, although the large number of observers (resulting from the use of primary care veterinary practice records) may have led to inconsistency, this does limit the chance of individual observer bias affecting the classification of HRI and means it is likely a more representative sample of the general UK dog population as primary care cases with HRI were included. Although this makes the study more generalisable, it is possible that these inclusion criteria may have selected for cases where HRI was not the true cause of clinical signs. However, the findings of increased brachycephalic risk for heatstroke were similar in Segev et al. (2015a) and Segev et al. (2015b) which had more specific inclusion criteria for dogs with more severe clinical signs of heatstroke.  Moreover, Davis et al. (2017) found that brachycephalic dogs developed hyperthermia more rapidly when exposed to heat stress, with significantly increased attempts at thermoregulation, compared to non-brachycephalic dogs. Although this study had methodology which used unnatural conditions compared to dogs presenting with HRI in veterinary practices, it does provide a standardisation of environment to ensure there is direct comparison between the different breeds and reducing the chance of differing environmental or exercise patterns being responsible for the increased risk.

A further limitation for consideration was that all the studies relied on breed classification to determine whether a dog was brachycephalic. Although Hall et al. (2020) identified skull shape as a risk factor to increase odds for HRI, the classification was based on recorded breeds and average skull morphology of the breed. The rest of the studies (Segev et al., 2015a; Segev et al., 2015b; and Davies et al., 2017) did not state their criteria for classification of breeds. No studies evaluated skull shape or degree of brachycephaly of the individual dogs studied as an independent variable to breed classification. There will be a wide variation of skull morphology present in dogs classified as a brachycephalic breed (Packer et al., 2015) especially as there will be variation in breed recording according to the observer, and not all animals classified as a particular breed will meet the breed standard for skull and muzzle conformation. This will especially impact the studies using clinical records to determine signalment. Further studies where skull to muzzle ratios are recorded would strengthen the evidence and reduce the risk of bias due to breed classification. However, some of this impact will be mitigated as it is thought that breed classifications tend to represent a minimum extreme for the breed (Knowler et al., 2019; and Packer et al., 2015), so it is likely that the evidence is still strong for brachycephalic dogs experiencing a higher risk of HRI.

It should also be noted that the studies analysed are from several different countries and include widely differing sample sizes, populations, and study designs. Although this means the results are not directly comparable, and therefore an overall prediction of odds or degree of increased risk is not possible, the population from the collective samples in these studies is felt to be generalisable to the population seen in veterinary practice, so provides meaningful evidence for a wide population of dogs.

The findings from these studies are felt to be highly relevant for dog welfare, especially given the current popularity of certain brachycephalic breeds such as French Bulldogs (O’Neill et al., 2018). Alongside this, there are concerns that skull and muzzle conformation are becoming more extreme in brachycephalic dogs (Knowler et al., 2019, Brachycephalic Working Group, 2020), which may further compound their risk of HRI. Raising awareness of the risk of HRI in brachycephalic dogs is important to help protect dogs from this potentially life-threatening condition. Efforts should be made to educate owners about the impacts exertional HRI (Segev et al., 2015a; and Hall et al., 2020) as well as the risk of environmental HRI, especially given the impacts of climate change which may lead to increased temperatures and extreme weather events in the UK and across the world (IPCC, 2022).

Overall, the studies provide moderate evidence to support brachycephaly as a risk factor in the development of HRI. Across all the cohort studies (Hall et al., 2020; Segev et al., 2015a; and Segev et al., 2015b), brachycephalic dogs were over-represented in presentations of HRI and in Davis et al. (2017), they were found to be more affected by exposure to heat stress. To provide stronger evidence, further study including systematic reviews with meta-analysis, impact of BOAS and specific investigations of skull morphology instead of breed classification would be beneficial. Additionally, it is also important to consider that HRI is a multifactorial condition and that other factors, such as body condition score and age, are also likely to be significant factors in the development of HRI in addition to skull morphology.

[…]