Observations and analysis of risk factors for GDV in a single institution: a case - control study

Objective To evaluate/ determine the risk factors for dogs presented with Gastric Dilation - Volvulus (GDV) to a referral veterinary centre and to compare the results with those currently reported in veterinary literature. The observational case - control study comprised a population of 215 dogs that attended the referral centre between 2000 and 2018. Medical records were searched for GDV and those matching the criteria were manually reviewed. A questionnaire was completed by the owners of the dogs involved in the study (both case and control populations) in order to gain further information that could be considered relevant to GDVs. In particular, information included the dogs ’ behaviour and the dogs daily activities. The study included 115 cases of GDV and 115 controls. Out of the GDV dogs 13% (15/115) were small breed dogs and the remaining 87% either large or giant breeds. with one study reporting a higher incidence during winter months 21, whereas another reported an increase in environmental temperature presented as a significant risk factor for GDV 16 . The association between GDV and physical activity remains controversial 15, 16. Recently, a study on canine gut microbiome showed that microbiome in GDV patients was altered by an expansion of a minor lineage and was associated with specific alleles of both innate and adaptive immunity genes. These associations are consistent with the hypothesis that immune genes may play a role in predisposition to GDV by altering the gut microbiome 22 . The purpose of this epidemiological survey was to determine whether some of the most commonly claimed risk factors for GDV exist in northern Italy and to ascertain whether there are differences when comparing similar studies conducted, to date, in other countries.


Introduction
Gastric Dilatation -Volvulus (GDV) is an acute and potentially fatal syndrome that mainly affects large deep-chested dogs, although occasionally it has been reported in small dogs and other species, including cats 1-5 guinea pigs 7, monkeys 8 and humans 9 .Dilatation is supposed to precede torsion, though it has been reported that gastric torsion episodes may arise without prior dilatation 9 . Symptoms predominate due to rapid accumulation of gases and fluids within the gastric lumen leading to distension, increasing intraluminal pressure and stomach rotation; those in turn lead to dyspnoea, varying degrees of hypovolemic and/or septic shock, ventilation-perfusion mismatch, systemic inflammatory response syndrome (SIRS), coagulation malfunction and multiple organ dysfunction syndrome (MODS) [10][11][12][13] , including ventricular arrhythmias and Disseminated Intravascular Coagulation (DIC) 14 . If left untreated, GDV can lead to severe impairment of venous return to the heart, diminished stroke volume, necrosis of the gastric wall and eventual fatality 15 .
Although the exact aetiology of GDV is still unknown, numerous predisposing risk factors have been described 12,16 . Male, large and giant pure breed dogs 17,18 , and deep chested dogs are at increased risk 19 . Other reported risk factors include increased age 18 , low Body Condition Score (BCS) 17 , nervous or a fearful temperament 17 , stressful events occurring within 8 hours prior to the GDV episode 12 , and a familial history of GDV 18 . Nutrition-related risk factors have also been identified, such as eating one meal per day, small food particles, rapid food ingestion 17,20 and eating only one type of food 14 . Some authors have suggested a relationship between GDV and weather conditions with one study reporting a higher incidence during winter months 21, whereas another reported an increase in environmental temperature presented as a significant risk factor for GDV 16 . The association between GDV and physical activity remains controversial 15, 16. Recently, a study on canine gut microbiome showed that microbiome in GDV patients was altered by an expansion of a minor lineage and was associated with specific alleles of both innate and adaptive immunity genes.
These associations are consistent with the hypothesis that immune genes may play a role in predisposition to GDV by altering the gut microbiome 22 .
The purpose of this epidemiological survey was to determine whether some of the most commonly claimed risk factors for GDV exist in northern Italy and to ascertain whether there are differences when comparing similar studies conducted, to date, in other countries.

Methods
The epidemiological case-control study was

Statistical Analysis
Statistical analysis, performed using statistical software R. Data was merged making non-dichotomous variables into dichotomous variables. For example, for the "food intake speed" parameter, the owner could choose between four answers: slow, normal, quick, voracious. The data was then grouped in slow/normal (Group 1) and fast/voracious (Group 2), using the Pivot tables. This procedure was carried out for all non-dichotomous variables (i.e. those variables for which the answer did not include a simple YES/NO but a scale of values). These variables were: the speed of food intake; temperament of the dog, type of feeding, position of the bowl, number of meals, type and frequency of physical activity, presence of episodes of vomiting and/or diarrhoea, tendency to eat foreign bodies, tendency to consume grass, stay in kennels, and whether it was used to long journeys.
Once all dichotomous variables were rendered, each parameter was entered in 2x2 contingency tables.
For some variables, several 2x2 tables were created, for example 4 tables were created for physical activity including the possibility to carry out physical activity, the possibility to carry out physical activity before the meal, the possibility to carry out physical activity after the meal, possibility to perform physical activity before and after the meal.
From the 2x2 contingency tables the following values were calculated: OR, ES, OR MIN and OR MAX.
OR is a statistical value that represents the relationship between sick and exposed subjects and the healthy and exposed subjects. It is calculated as (a/c)/(b/d), ie (cases/non-cases) exposed/(cases/non-cases) not exposed. (TABLE 2). no difference between exposed or non-exposed subjects to risk factors). If the confidence interval does not contain the value 1, the null hypothesis is rejected, i.e.
the alternative hypothesis is accepted, which states that there is a difference between exposed and non-exposed subjects to the risk factor, so that the OR value is significant (         Owners were asked if their dogs were used to ingest accidentally or spontaneously things other than food ( grass or foreign body). It has been seen that dogs that tend to ingest grass were more predisposed to develop GDV than dogs that do not have this feeding behaviour.
The calculated values (OR = 2.92, CL95% = 1.61-5.30 and P = 0.0006) testify that the ingestion of grass represents a risk factor for the GDVAbout foreign bodies a recent paper shows that there is a correlation between foreign body and development of GDV (24)  Despite the complexity of the risk factors, it is possible to reduce the incidence of GDV in high-risk breeds by observing some precautions 25 . It is not recommended to feed the dog only once a day 25 . They should be fed with small amounts of food per meal and have more meals per day (at least 2) 18 . It would also be useful to reduce the speed of recruitment 14,25 , as slowing the speed of intake has a preventative effect 7 .
Feeding with dry commercial food alone may not be a well suited choice for dogs at risk 20 adding home food to the commercial diet may be useful for preventing the syndrome 14 . It is advisable to avoid administering only a commercial crunchy diet to those at risk 14 .
There are still doubts about the role of physical activity; according to some authors, the intense physical activity should be reduced in the two hours after the meal 14 . According to others, moderate daily physical activity and moderate post-prandial exercise appears to be beneficial for the reduction of risk 20 .
Owners of dogs of the breeds more predisposed to the disease should also be advised to restrict access to foreign material as much as possible, as the presence of a FB at the gastric level increases the possibility of developing GDV.
Regarding other aspects, it would certainly be useful for purebred clubs to work closely with researchers and veterinary surgeons to better identify specific morphometric and genetic factors that predispose certain dogs to the development of GDV 14 .

This Study Contains Several Limitations
In a case-control study, the case group and the control group are selected on the basis of the outcome (ie, having the disease of interest compared to not having the disease) and comparing the two groups with respect to their previous exposure frequencies to possible risk factors. Recruitment of subjects is specifically chosen by groups of subjects with or without the disease of interest 26 .
Conversely there are also advantages of this type of study. It is inexpensive, useful and effective for investigating a infrequently occurring disease. The clinician was also able to access a lot data from a single subject. Interviews with one owner granted information on signalling, eating habits and lifestyle.
Nonetheless, the case-control study is very sensitive to BIAS (or distortion). This term means the difference between the estimate obtained from the sample and the true characteristic of the population. To avoid a distortion in the identification and quantification of risk factors, it would be useful to match the cases with controls that are as similar as possible (for example the same age, sex and breed) using a procedure called "matching" 26 . However, in this study this type of procedure was not used and the controls were chosen on the basis of other criteria, as explained below.
Furthermore it is the opinion of the authors that increasing the sample size might change final results.

Conclusion
In this study several risk factors have been identified for GDV development. These include; large and giant dogs, purebred dogs, dogs over 3 years old, a weight>30 kg, a diet which consists of commercial dry food, the ingestion of foreign bodies and grass, a history of episodes of diarrhoea and/or vomiting, exercise more than once a day and both before and after the meal, the consumption of a single daily meal and the speed of food intake.
On the contrary, neither sex nor the status of the subject (integer/neutered), temperament, BCS and the habit of long journeys or to stay in pensions or kennels are to be considered risk factors.
This study concludes that the risk of GDV in certain breeds and in certain conditions remains high.
Since GDV continues to be an extremely current disease it is important that risk factors are understood. Although safe method to prevent the disease is preventive gastropexy [27][28][29][30][31][32] .

Competing Interests
The authors declare that they have no competing interests.

Availability of Data and Materials
All data generated or analysed during this study are included in this article.

Not applicable
Ethics Approval and Consent to Participate:

Funding
The authors declare that there were no funding and support for the study.