Owing to their virtually incomparable olfactory apparatus and the mutual loving relationship with man, the use of dogs for assisting humans in many activities has become commonplace. Dogs have been used for long for livestock herding, hunting and pulling. More recently, they have been employed for servicing or assisting people with disabilities, for rescuing, for pet therapy and, last but not least, for detecting a vast array of volatile organic compounds related to drugs, narcotics, explosives and foods. Although cancer detection seems the most distinguished use of “man’s best friends” in science and medicine, increasing emphasis is being placed on their capacity to perceive chemical changes or human expressions associated with harmful, even life-threating, blood glucose variations. The evidence available in the current scientific literature attests that diabetes alerting dogs (DADs) have a heterogeneous efficiency for warning owners of episodes of hypoglycemia or hyperglycemia, with sensitivities and specificities ranging between 0.29–0.80 and 0.49–0.96, respectively. Although the adoption of DADs seems effective for improving the quality of life of many diabetics patients, some important drawbacks can be highlighted. These typically include adoption and keeping expenditures, lack of certification or accreditation of dog providers, poor harmonization of training procedures, significant inter-breed, intra-breed and intra-dog variabilities, wide-ranging alert behaviors, ability of owners to identify dog’s alerts, as well as lack of quality assessment of a dog’s “diagnostic” performance. Overcoming many of these limitations shall probably make DADs more efficient tools for improving diabetes management.
Domestic dog, a member of the genus Canis (canines), was the first animal to be domesticated by humans, and is now universally regarded as the “man’s best friend” . Besides being widely used as pets, dogs also help humans in many activities, thus assuming the conventional nickname of “working dogs”. The most common tasks performed by dogs include activities as old as their domestication (e.g. livestock herding, hunting and pulling), more recent tasks (servicing or assisting people with various forms of disability, for rescuing or for pet therapy), along with a series of activities directly attributable to their incomparable sniffing capacity.
The canine olfactory system, which is probably the most developed after that of wolves and bears , gives them the capacity to detect even negligible amounts of volatile organic compounds (VOCs) related to drugs, narcotics, explosives and foods. Hence, it is commonplace to see dogs involved in biosecurity, drug enforcement, food security, prevention and containment of bio- and agro-terrorism. Ideally, dogs’ efficiency to search target odors make them excellent “tools” for rapid and efficient target recognition. As a matter of fact, canine olfactory perception is vastly more sensitive than the human sense of smell. This is mostly due to a number of genetic, anatomic and physiological differences, allowing dogs to more efficiently record and cognitively analyze chemosensory information present in the environment . Several lines of evidence now attest that the genetics of olfactory receptor apparatus has been subjected to considerable expansion and contraction throughout mammal evolution, thus resulting in an enormous molecular, biochemical, anatomic and function heterogeneity . The overall area including olfactory cells is approximately 5 cm2 in human nose, whilst is over 20-fold larger in dogs (i.e. usually comprised between 150 and 170 cm2), so that the final number of olfactory receptors is >20-fold higher in dogs than in humans (i.e. ~5 vs. ~220 million) .
This virtually incomparable smelling capacity has contributed to broaden the use of dogs to many detection applications, including disease recognition. Albeit the most known relevance of these friendly animals in science and medicine is cancer detection [4, 5], increasing emphasis is now being placed on their capacity to perceive chemical (and perhaps expression) changes in human body often associated with remarkable variations of blood glucose concentration. With specific training, a dog may hence be able to warn its owner on both hypoglycemic and hyperglycemic episodes. The relative short story of diabetes alerting dogs (DADs) probably commenced in the 2003 when Armstrong, a yellow Labrador Retriever, received an outstanding award (and was hence included in the Guinness Book of World Records) for being able to sniff subtle chemical changes, and then alerting his type 1 diabetic owner on an ongoing hypoglycemic episode (Figure 1) .
Some time ago we published an article aimed at reviewing scientific evidence on the use of DADs for a cost-effective approach of diabetes monitoring, which was almost entirely based on anecdotal evidence, non-human studies or case reports . Some years afterward, a larger number of interesting studies were published, which promoted us to provide a narrative, critical update on this matter.
Published evidence on glycemic alerts
The very first study providing some reliable evidence that dogs may be able to warn their diabetic owners about potentially harmful glycemic states was published by Wells et al., in 2008. The authors enrolled 212 type 1 diabetes owning personal dogs (i.e. untrained to detect blood glucose variations) . Each participant was asked to fill in a specific questionnaire for collecting data on dog response to hypoglycemic episodes. Cumulatively, 65.1% of diabetic subjects indicated that the dogs displayed some behavioral reactions during at least one hypoglycemic event, whilst approximately one third of dogs reacted to the vast majority of episodes. Notably, 33% of responders indicated that their dogs displayed some unusual behavior even before the owner recognized that they were suffering from a hypoglycemic state. The most frequent dogs’ alert behaviors were vocalizations (61%), licking (49%), staring intently at owner’s face (41.3%), nuzzling (41%) or jumping (30%). This first important evidence lead the way to publication of other ensuing studies, which were more focused on the role of DADs, rather than on “normal” pets, for warning owners on potentially dangerous variations of blood glucose concentration.
Rooney et al. enrolled 17 diabetic subjects (one with type 1 diabetes and the remaining 16 with type 2 diabetes), who had owned a DAD (six Labrador Retrievers, two Labradoodles, two Labrador Retriever/Golden Retriever cross, and one Golden Retriever, one Collie Cross, one Poodle, one Yorkshire Terrier, one Lurcher and one Cocker Spaniel) for a mean period of 1.9 years . Glycemic alerts were identified by blood glucose readings, whilst the exact observation period was not declared (i.e. ≥1 month). Notably, the sensitivity of DADs for alarming their owners to glucose measurements lying outside the target range was 0.80 (95% confidence interval [95% CI], 0.44–0.97), whilst data on specificity were unavailable in this study. Interestingly, glycated hemoglobin (HbA1c) values displayed a modest and non-significant reduction after owning the dog (i.e. 62.7 vs. 61.2 mmol/mol), whilst the rate of blood glucose readings below the target range was remarkably lower after owing DADs (odds ratio [OR], 0.24; 95% CI, 0.17–0.35). Finally, 75% of patients declared that they felt more independent after living with the DAD.
Dehlinger et al. carried out another original investigation using three DADs, which were trained to identify hypoglycemia from a series of sterile cotton swabs rubbed on the skin of three type 1 diabetic patients collected over a spectrum of 24 various glycemic states (half of which were hypoglycemic) . The cumulative diagnostic accuracy, sensitivity and specificity of the three DADs for identifying hypoglycemic samples were 0.55, 0.56 and 0.54, respectively.
In an ensuing study, Hardin et al. assessed the ability of six, 6-month trained DADs (two Labrador Retrievers and one Flat Coated Retriever, one German Shepherd, one Siberian Husky mix and one Spaniel mix) for detecting hypoglycemia in human perspiration samples . Dogs were specifically trained to signal hypoglycemic samples by sitting in front of, or pushing the can containing the sample (the dogs were challenged with seven human perspiration samples, one of which was collected during a hypoglycemic state). The cumulative diagnostic accuracy, sensitivity and specificity of DADs for detecting perspiration samples collected during hypoglycemic episodes were 0.93, 0.78 and 0.96, respectively.
Los et al. performed an exploratory study, including eight subjects with type 1 diabetes, owning a DAD (five Labrador or Labrador mix and one Golden Doodle, one Border Collie and one Saint Bernard) . Glycemic alerts were identified by using continuous glucose monitoring, during a 3-month period. The cumulative sensitivity of DADs for detecting hypoglycemia was 0.36, whilst no information was reported about the specificity (the false positive rate was 14.5). Notably, the combination of DAD alert and onset of subjective symptoms increased by nearly five-fold the chance of being warned about hypoglycemic episodes compared to the presence of subjective symptoms alone (i.e. 62% vs. 12%). Nevertheless, it was also shown that continuous glucose monitoring would have warned the owner prior to the DAD in 73% of cases. Interestingly, the median cost of adopting a DAD was estimated to be between 5000 and 15,000 US $, whilst the home-based training program cost 2000 US $.
Gonder-Frederick et al. carried out another study including 14 diabetics (seven adults with type 1 diabetes and seven youth with type 1 diabetes) who had DADs (all Labrador Retrievers) placed in their home for a minimum of 6 months . Glycemic alerts were identified by using continuous glucose monitoring devices. Overall, 2278 hypoglycemic or hyperglycemic events could be recorded over a median period of 29 days. The DADs cumulative sensitivity and specificity were 0.29 (0.36 for hypoglycemia and 0.26 for hyperglycemia, respectively) and 0.66, thus yielding to a cumulative diagnostic accuracy of 0.48. The accuracy of DADs was characterized by a high inter-dog variability.
Short afterwards, the same team of authors performed another study based on 18 type 1 diabetic subjects who had most DADs (all Labrador Retrievers) in their home for ~3 months . Glycemic alerts were identified by using glucose monitoring devices. Overall, 3006 hypoglycemic or hyperglycemic events could be recorded over a median period of 134 days. The DADs cumulative sensitivity and specificity were 0.57 (0.59 for hypoglycemia and 0.56 for hyperglycemia, respectively) and 0.49, thus yielding to a cumulative diagnostic accuracy of 0.54. Even in this study a highly variable inter-DAD accuracy was observed.
Taken together, the available published studies analyzing dog’s efficiency for alerting their owners on episodes of hypoglycemia or hyperglycemia seems hence rather heterogeneous, with sensitivities and specificities ranging between 0.29–0.80 and 0.49–0.96, respectively (Table 1). Moreover, a substantial inter-dog variability has been observed in the majority of the studies.
|Study||Sample size||Observation period||Outcome||Performance|
|Rooney et al., 2013 ||17 Subjects (1 type and 16 type 2 diabetes)||≥1 Month||Detecting hyperglycemia||0.80 Sensitivity|
|Dehlinger et al., 2013 ||3 Dogs each challenged with 24 Human cotton swabs (12 collected during hypoglycemia)||–||Detecting hypoglycemia||Accuracy, 0.55 (0.56 sensitivity and 0.54 specificity)|
|Hardin et al., 2015 ||6 Dogs each challenged with 7 Human perspiration samples (1 collected during hypoglycemia)||–||Detecting hypoglycemia||Accuracy, 0.93 (0.78 sensitivity and 0.96 specificity)|
|Los et al., 2017 ||8 Type 1 diabetics||3 Months||Detecting hypoglycemia||Sensitivity, 0.36|
|Gonder-Frederick et al., 2017 ||14 Subjects (7 with type 1 and 7 With type 2 diabetes)||Median, 29 Days||Detecting both hypo- and hyperglycemia||Cumulative accuracy, 0.48 (0.29 sensitivity and 0.66 specificity)|
|Gonder-Frederick et al., 2017 ||18 Type 1 diabetics||Median, 134 Days||Detecting both hypo- and hyperglycemia||Cumulative accuracy, 0.54 (0.57 sensitivity and 0.49 specificity)|
Additional evidence on glycemic control and quality of life
Gonder-Frederick et al. carried out a small exploratory survey, which was completed by 36 DAD owners (i.e. 13 adults with type 1 diabetes and 23 relatives of children with type 1 diabetes), aimed at identifying glycemic control and quality of life improvements after owing the dog . Overall, a significantly lower rate of severe (no frequency reported; p=0.039) and moderate (no frequency reported; p=0.02) hypoglycemic episodes was declared by responders after owing a DAD, whilst the HbA1c values were also found to be lower after DAD placement (no values reported; p<0.001).
Petry et al. surveyed 135 subjects (72 diabetic patients and 63 parents of children with diabetes) with the aim of evaluating owners’ perceptions about DADs (nearly half were Labrador Retrievers) or their DAD’s capacity of detecting potentially harmful glucose levels . More than half of responders stated that the dog was capable of frequently identifying (i.e. more than three times per week) low glucose values when the owner was unaware of them, whilst ~80% of responders felt that the dog may have saved his/her life at least once. Finally, the vast majority of responders stated that the dog was very or extremely useful for managing diabetes. According to the replies to this survey, DAD also had a substantial favorable impact on perceived quality of life, by reducing the frequency of hospitalization, need of assistance from other persons, as well as by decreasing the alleged risk of accidents while driving a car.
Lundqvist et al. carried out another longitudinal interventional study including 55 owner-and-dog pairs, 20 of which were DADs . A specific questionnaire aimed at recording several health-related quality of life and activities was administered before and 3 months after adopting the dog. Notably, diabetic owners of DADs reported a higher health-related quality of life compared to that of the general population, but no significant improvement could be recorded after living with the dog. The analysis of the entire study population also revealed that physical activity level, health-related quality of life and well-being had all improved significantly after owing the dog.
What are diabetes alert dogs really perceiving?
Unlike other human diseases, it is likely that dogs may be able to perceive human glycemic variations not only using their exceptional olfactory apparatus, but also by integrating perceptions coming from other senses, mainly from their sight.
Indeed, the canine olfactory apparatus plays a pivotal role for identifying several organic compounds (i.e. the so-called “volatilome”) generated during diseased conditions (especially cancer, seizures, infectious and psychiatric diseases), as earlier proven in many published studies [18, 19]. With the aim of replicating dogs’ detection ability, Siegel et al. carried out an interesting study, in which VOCs collected from the breath of 52 type 1 diabetics were analyzed with gas chromatography/mass spectrometry (GC/MS) . The ensuing linear discriminant analysis showed that hypoglycemic episodes were associated with highly characteristic signatures, displaying 0.91 sensitivity and 0.84 specificity. In another recent study, Saidi et al. collected breath samples from six diabetic patients, and compared their profiles with those of non-diabetic subjects, using a GC/MS technique . The resulting radar plots were clearly different in diabetics, displaying increased concentrations of benzaldehyde, toluene, methane, aniline and carbonochloridic acid. Other studies were capable of identifying a number of putative VOCs in the exhaled breath during hypoglycemia, which may then be sniffed by dogs. These basically include isoprene , along with a vast array of volatile substances that are typically increased during hypoglycemia such as ketones (e.g. acetone and acetaldehyde), hydrocarbons (e.g. ethane, pentane and isoprene), acetaldehyde, methanol and ethanol . This preliminary evidence leads the way for planning further investigation aimed at exploring the association between dogs’ alerts and variation of measured volatile organic in exhaled breath or in other biological fluids (e.g. sweat or saliva) during hypoglycemic episodes and, especially, in the context of severe hyperglycemia, for which no reliable studies have been published so far .
It is now unquestionable that dogs can read human facial expressions and postures. In particular, recent studies showed that dogs are very sensitive to emotional expressions conveyed by the human face , emotional vocalizations  and sudden change of attention or posture , thus being even able to discriminate between positive and negative emotions . It is hence conceivable that the dog’s cognitive apparatus may be able to elaborate and integrate a number of different owners’ olfactory and visual signals, which may finally help the pet recognizing harmful clinical conditions (Figure 2).
Although most of us would agree that DADs have a conjectural potential for improving the quality of life (both clinical and social) of many diabetic patients (Table 1), at least those who are not allergic and can love and respect them, some important drawbacks can be easily identified for using these pets as diagnostic aids (Table 2).
|–||Cost for adopting and keeping the dog|
|–||Certification or accreditation of dogs’ providers|
|–||Harmonization of training procedures|
|–||Inter-breed, intra-breed and intra-dog variability|
|–||Heterogeneity of dogs’ alert behaviors|
|–||Capability of owners to identify altered behaviors|
|–||Quality assurance and verification|
The first issue relates to the cost for adopting and keeping these “special” dogs. The current expenditure of a DAD varies widely, from approximately one to several thousands of US $. For example, an US organization applies an one-time fee of 1600 US $ for providing a completely trained DAD , whilst others have prices ranging from 8500 US $  to over 20,000 US $ , or target the cost depending on individual needs . Occasionally, DADs may also be provided by not-for-profit organizations, but these are only sporadic cases in the growing DAD industry. No country has made the adoption of DADs refundable so far, and hence purchasing and keeping the pet may pose a considerable economic burden on some diabetic subjects and their families. Then, no official certification or accreditation exists for associations, organizations or even single trainers of DADs, so that it cannot be taken for granted that dogs have been appropriately or accurately trained to detect severe blood glucose fluctuations.
As a mainstay of diagnostic testing, the “calibration” procedure is critical for obtaining accurate and, especially, comparable clinical information across different settings . Albeit the DAD’s training procedures used in the different studies were mostly based on positive reinforcement and standard training stimuli (e.g. usually with food), thus avoiding coercion-based strategies, each trainer exploits a specific method and, understandably, dogs are not laboratory instrumentation, so that they will manifest a highly subjective reaction to training and stimuli. Therefore, although standardizing training procedures seems a reachable goal, harmonization of individual dog’s behaviors will probably remain an unmet target .
A large inter-breed, intra-breed and even intra-dog variability of olfactory capacity has been clearly described. Irrespective of the huge heterogeneity in the total number of olfactory receptors across different breeds, ranging between 120 million in Dachshunds to over 300 million in Bloodhounds , Tacher et al. analyzed the sequence of as many as 16 olfactory receptor genes in 95 dogs of 20 different breeds , and found high molecular heterogeneity, leading to significant amino acid variations. This perhaps contributes to justify the different odor discrimination performance observed across different breeds (e.g. wolves and scent dogs such as German Pointers, Basset Hounds and Beagles, perform much better than short-nosed dogs or non-scent breeds), as well as among dogs belonging to the same breed . These findings were then replicated by Lesniak et al., who also found that some polymorphisms in certain genes may have a considerable impact on canine odor sensing skills [38, 39]. Importantly, circadian and locomotors behaviors , along with metabolic variations , also induce significant variations of olfactory perceptions. It is hence predictable that the individual odor discrimination performance of dogs might display a substantial within-day and day-to-day variability.
The heterogeneity of dogs’ alert behaviors, as well as the capability of a dog’s owner to identify altered behaviors, are additional limitations. The current published studies have listed a large number of potential alert behaviors, including barking, whining, hiccupping, licking, yawning, scratching, biting, nudging, nuzzling, jumping or staring intently at the owner’s face [8, 9]. All these are commonplace dog activities, and it will not be easy for the owner to distinguish whether an unusual behavior of his/her dog has been triggered by severe glycemic derangements or by another environmental circumstance. Then, even the owners themselves may not have enough skill, vigilance or be ready to capture these alert behaviors, so that there is a tangible risk that dog’s warning may go overlooked.
From a genuine technical perspective, to which we are especially attached as we are both clinical biochemists, quality assessment of dogs’ olfactory performance is still an unexplored domain. Quality assurance and verification [42, 43], along with regular use of quality indicators , are now essential parts of total quality in laboratory diagnostics. Albeit the translation of these concepts to dogs’ olfactory performance seems not easy and straightforward, because the “quality control material” cannot be different from the owners themselves (e.g. each human has a virtually unique smell fingerprint, to which the pet is accustomed) , poor individual performance of the dog, or progressive decay of odor sensing skill over time, may then generate an unacceptable rate of false negative and false positive alerts, which may either falsely reassure the owner about a normal (i.e. euglycemic) state or generate unnecessary and unjustified concern, respectively. It may hence be necessary to define standard operating procedures to be used by owners for continuous monitoring of their dog’s alert performance, more or less like routine checking the accuracy of glucose meters or continuous glucose monitoring devices. A simple approach may be based on the use a diary for correlating glucose fluctuations with dog’s behaviors, especially focused on registering episodes of hypoglycemia and hyperglycemia. The owner’s personal physician (or diabetologist) may then calculate and number of false positive and false negative alerts, benchmarking them with their dog’s previous accomplishments or with other DADs’ performance. Nevertheless, like other diagnostic domains, this second aspect shall first need to develope reliable quality indicators  and informatics tools, where a dog’s olfactory performance could be recorded in a consistent and standardized manner, and then be compared with benchmark performances .
In conclusion, the current published evidence attests that DADs may contribute to provide some undeniable health  and psychological  benefits to diabetic people, though their efficiency for detecting harmful, even life-threatening, blood glucose variations remains questionable. Some important case reports have demonstrated that “man’s best friends” have actually saved the lives of some diabetic subjects, especially when immediate glucose monitoring was unavailable . Therefore, it is the authors’ opinion that albeit DADs will never replace self-testing, especially glucose continuous monitoring, their gracious and affectionate presence may still be a valuable aid for promoting healthier lifestyles, for improving the overall quality of life in diabetes and, under special circumstances, for warning owners on dangerous glucose fluctuations.
This article is dedicated to Armstrong (Figure 1) and to all the other dogs engaged in saving human lives.
Author contributions: All the authors have accepted responsibility for the entire content of this submitted manuscript and approved submission.
Research funding: None declared.
Employment or leadership: None declared.
Honorarium: None declared.
Competing interests: The funding organization(s) played no role in the study design; in the collection, analysis, and interpretation of data; in the writing of the report; or in the decision to submit the report for publication.
1. Udell MA, Wynne CD. A review of domestic dogs’ (Canis familiaris) human-like behaviors: or why behavior analysts should stop worrying and love their dogs. J Exp Anal Behav 2008;89:247–61.10.1901/jeab.2008.89-247Search in Google Scholar PubMed PubMed Central
2. Hughes GM, Boston ES, Finarelli JA, Murphy WJ, Higgins DG, Teeling EC. The birth and death of olfactory receptor gene families in mammalian niche adaptation. Mol Biol Evol 2018;35:1390–406.10.1093/molbev/msy028Search in Google Scholar PubMed PubMed Central
7. Lippi G, Cervellin G, Dondi M, Targher G. Hypoglycemia alert dogs: a novel, cost-effective approach for diabetes monitoring? Altern Ther Health Med 2016;22:14–8.Search in Google Scholar
9. Rooney NJ, Morant S, Guest C. Investigation into the value of trained glycaemia alert dogs to clients with type I diabetes. PLoS One 2013;8:e69921.10.1371/journal.pone.0069921Search in Google Scholar PubMed PubMed Central
10. Dehlinger K, Tarnowski K, House JL, Los E, Hanavan K, Bustamante B, et al. Can trained dogs detect a hypoglycemic scent in patients with type 1 diabetes? Diabetes Care 2013;36:e98–9.10.2337/dc12-2342Search in Google Scholar PubMed PubMed Central
11. Hardin DS, Anderson W, Cattet J. Dogs can be successfully trained to alert to hypoglycemia samples from patients with type 1 diabetes. Diabetes Ther 2015;6:509–51.10.1007/s13300-015-0135-xSearch in Google Scholar PubMed PubMed Central
12. Los EA, Ramsey KL, Guttmann-Bauman I, Ahmann AJ. Reliability of trained dogs to alert to hypoglycemia in patients with type 1 diabetes. J Diabetes Sci Technol 2017;11:506–12.10.1177/1932296816666537Search in Google Scholar PubMed PubMed Central
13. Gonder-Frederick LA, Grabman JH, Shepard JA. Diabetes alert dogs (DADs): an assessment of accuracy and implications. Diabetes Res Clin Pract 2017;134:121–30.10.1016/j.diabres.2017.09.009Search in Google Scholar PubMed PubMed Central
14. Gonder-Frederick LA, Grabman JH, Shepard JA, Tripathi AV, Ducar DM, McElgunn ZR. Variability of diabetes alert dog accuracy in a real-world setting. J Diabetes Sci Technol 2017;11:714–9.10.1177/1932296816685580Search in Google Scholar PubMed PubMed Central
15. Gonder-Frederick L, Rice P, Warren D, Vajda K, Shepard J. Diabetic alert dogs: a preliminary survey of current users. Diabetes Care 2013;36:e47.10.2337/dc12-1998Search in Google Scholar PubMed PubMed Central
16. Petry NM, Wagner JA, Rash CJ, Hood KK. Perceptions about professionally and non-professionally trained hypoglycemia detection dogs. Diabetes Res Clin Pract 2015;109:389–96.10.1016/j.diabres.2015.05.023Search in Google Scholar PubMed PubMed Central
17. Lundqvist M, Levin LÅ, Roback K, Alwin J. The impact of service and hearing dogs on health-related quality of life and activity level: a Swedish longitudinal intervention study. BMC Health Serv Res 2018;18:497.10.1186/s12913-018-3014-0Search in Google Scholar PubMed PubMed Central
19. Angle C, Waggoner LP, Ferrando A, Haney P, Passler T. Canine detection of the volatilome: a review of implications for pathogen and disease detection. Front Vet Sci 2016;3:47.10.3389/fvets.2016.00047Search in Google Scholar PubMed PubMed Central
20. Siegel AP, Daneshkhah A, Hardin DS, Shrestha S, Varahramyan K, Agarwal M. Analyzing breath samples of hypoglycemic events in type 1 diabetes patients: towards developing an alternative to diabetes alert dogs. J Breath Res 2017;11:026007.10.1088/1752-7163/aa6ac6Search in Google Scholar PubMed
21. Saidi T, Zaim O, Moufid M, El Bari N, Ionescu R, Bouchikhi B. Exhaled breath analysis using electronic nose and gas chromatography–mass spectrometry for non-invasive diagnosis of chronic kidney disease, diabetes mellitus and healthy subjects. Sens Actuators B Chem 2018;257:178–88.10.1016/j.snb.2017.10.178Search in Google Scholar
22. Neupane S, Peverall R, Richmond G, Blaikie TP, Taylor D, Hancock G, et al. Exhaled breath isoprene rises during hypoglycemia in type 1 diabetes. Diabetes Care 2016;39:e97–8.10.2337/dc16-0461Search in Google Scholar PubMed
23. Qiao Y, Gao Z, Liu Y, Cheng Y, Yu M, Zhao L, et al. Breath ketone testing: a new biomarker for diagnosis and therapeutic monitoring of diabetic ketosis. Biomed Res Int 2014;2014:869186.10.1155/2014/869186Search in Google Scholar PubMed PubMed Central
24. Minh Tdo C, Blake DR, Galassetti PR. The clinical potential of exhaled breath analysis for diabetes mellitus. Diabetes Res Clin Pract 2012;97:195–205.10.1016/j.diabres.2012.02.006Search in Google Scholar PubMed PubMed Central
25. Siniscalchi M, d’Ingeo S, Quaranta A. Orienting asymmetries and physiological reactivity in dogs’ response to human emotional faces. Learn Behav 2018 Jun 19. DOI:10.3758/s13420-018-0325-2. [Epub ahead of print].10.3758/s13420-018-0325-2Search in Google Scholar PubMed
26. Siniscalchi M, d’Ingeo S, Fornelli S, Quaranta A. Lateralized behavior and cardiac activity of dogs in response to human emotional vocalizations. Sci Rep 2018;8:77.10.1038/s41598-017-18417-4Search in Google Scholar PubMed PubMed Central
27. Schwab C, Huber L. Obey or not obey? Dogs (Canis familiaris) behave differently in response to attentional states of their owners. J Comp Psychol 2006;120:169–75.10.1037/0735-7036.120.3.169Search in Google Scholar PubMed
28. Albuquerque N, Guo K, Wilkinson A, Savalli C, Otta E, Mills D. Dogs recognize dog and human emotions. Biol Lett 2016;12:20150883.10.1098/rsbl.2015.0883Search in Google Scholar PubMed PubMed Central
33. Myers GL, Miller WG. The roadmap for harmonization: status of the International Consortium for Harmonization of Clinical Laboratory Results. Clin Chem Lab Med 2018;56:1667–72.10.1515/cclm-2017-0907Search in Google Scholar PubMed
34. Teodoro-Morrison T, Diamandis EP, Rifai N, Weetjens BJ, Pennazza G, de Boer NK, et al. Animal olfactory detection of disease: promises and pitfalls. Clin Chem 2014;60:1473–9.10.1373/clinchem.2014.231282Search in Google Scholar PubMed
35. Padodara RJ, Jacob N. Olfactory sense in different animals. Ind J Vet Sci 2014;2:1–14.Search in Google Scholar
36. Tacher S, Quignon P, Rimbault M, Dreano S, Andre C, Galibert F. Olfactory receptor sequence polymorphism within and between breeds of dogs. J Hered 2005;96:812–6.10.1093/jhered/esi113Search in Google Scholar PubMed
37. Polgár Z, Kinnunen M, Újváry D, Miklósi Á, Gácsi M. A test of canine olfactory capacity: comparing various dog breeds and wolves in a natural detection task. PLoS One 2016;11:e0154087.10.1371/journal.pone.0154087Search in Google Scholar PubMed PubMed Central
38. Lesniak A, Walczak M, Jezierski T, Sacharczuk M, Gawkowski M, Jaszczak K. Canine olfactory receptor gene polymorphism and its relation to odor detection performance by sniffer dogs. J Hered 2008;99:518–27.10.1093/jhered/esn057Search in Google Scholar PubMed
39. Carskadon MA, Saletin JM, Van Reen E, Bartz A, Hart C, Raynor H, et al. Circadian influences on smell and taste detection thresholds: preliminary results from adolescents. Sleep 2015;38:213–28.Search in Google Scholar
41. Palouzier-Paulignan B, Lacroix MC, Aimé P, Baly C, Caillol M, Congar P, et al. Olfaction under metabolic influences. Chem Senses 2012;37:769–97.10.1093/chemse/bjs059Search in Google Scholar PubMed PubMed Central
44. Sciacovelli L, Panteghini M, Lippi G, Sumarac Z, Cadamuro J, Galoro CA, et al. Defining a roadmap for harmonizing quality indicators in Laboratory Medicine: a consensus statement on behalf of the IFCC Working Group “Laboratory Error and Patient Safety” and EFLM Task and Finish Group “Performance specifications for the extra-analytical phases”. Clin Chem Lab Med 2017;55:1478–88.10.1515/cclm-2017-0412Search in Google Scholar PubMed
45. Penn DJ, Oberzaucher E, Grammer K, Fischer G, Soini HA, Wiesler D, et al. Individual and gender fingerprints in human body odour. J R Soc Interface 2007;4:331–40.10.1098/rsif.2006.0182Search in Google Scholar PubMed PubMed Central
46. Sciacovelli L, Lippi G, Sumarac Z, West J, Garcia Del Pino Castro I, Furtado Vieira K, et al. Quality indicators in laboratory medicine: the status of the progress of IFCC Working Group “Laboratory Errors and Patient Safety” project. Clin Chem Lab Med 2017;55:348–57.10.1515/cclm-2016-0929Search in Google Scholar PubMed
47. Lippi G, Sciacovelli L, Simundic AM, Plebani M. Innovative software for recording preanalytical errors in accord with the IFCC quality indicators. Clin Chem Lab Med 2017;55:e51–3.10.1515/cclm-2016-1138Search in Google Scholar PubMed
48. Mubanga M, Byberg L, Nowak C, Egenvall A, Magnusson PK, Ingelsson E, et al. Dog ownership and the risk of cardiovascular disease and death – a nationwide cohort study. Sci Rep 2017;7:15821.10.1038/s41598-017-16118-6Search in Google Scholar PubMed PubMed Central
©2019 Walter de Gruyter GmbH, Berlin/Boston