Greyhounds given dental implants then killed at Melbourne Dental School

Mellati E, Chen S, Davies H, Fitzgerald W, Darby I 2015. ‘Greyhound –Healing of Bio-Oss® grafted marginal gaps at implants placed into fresh extraction sockets of incisor teeth in dogs: a study on the effect of submerged vs. non-submerged healing’. Clinical Oral Implants Research. 26: 553–562. doi: 10.1111/clr.12442.


Tooth implantation surgery usually requires that implants be initially covered with a flap of tissue to protect the site from infection, and to avoid structural complications. This ‘submerged’ method, requires another procedure to cut and remove the covering skin at a later time. A number of recent experiments, however, illustrate that, now, implants can be installed without this process. The authors of this study sought to compare the tissue and bone structure of the ‘submerged’ and ‘non-submerged’ procedures by analysing bone and tissue samples taken from greyhounds. In order to remove these samples the dogs were killed.

The Experiment

Six healthy female greyhounds of roughly 30kg each and aged between 1 and 2 years were used in this experiment. Prior to induction, each dog was sedated, a 22g catheter was inserted into the foreleg and then general anaesthesia was induced. Local anaesthesia was then provided to the facial region.

Incisions were made along the gums in front of the teeth so the tissue could be peeled back and the buccal bone (the bone in which teeth sit) was exposed. Two upper teeth (the maxillary incisors) were then removed from each dog using a luxator (a specially designed periodontal ligament knife ) to cut and forceps to pull.

The buccal bone was cut following the implant manufacturer’s specifications. Once the implants were installed the gap between the implant and surrounding bone walls was measured.

A coin toss was used to determine which implant would be submerged; a ‘closure screw’ and a ‘healing abutment’ were then installed at this implant. To encourage bone regrowth, the gaps between the implant and bone were then filled with a bone substitute (derived from cow) and covered with a ‘resorbable collagen membrane’ (derived from pig). The soft tissue below each implant was then cut with three incisions so as to become a ‘flap’. This flap was pulled to completely cover the ‘submerged’ implant and to cover only the buccal bone and “filling” of the ‘non-submerged’ implant.

Post-operative care included administration of a non-steroidal analgesic, and an injection of antibiotics directly into the muscle. During the first two weeks dogs received daily inspections and three times a week had their teeth brushed to control plaque.

At three months all the dogs were killed with an injection of Lethabarb®. The front portion of the upper jaws was then surgically removed and sent to a laboratory to be prepared for analysis.


Overall, the study showed that the results of both the ‘submerged’ and ‘non-submerged’ procedures of tooth implantation were very similar. Still, a number of complications of potential importance are suggested: the mean distance between first bone and implant contact was slightly less in submerged groups; likewise, the vertical distance between the regenerated bone (buccal crest) and bone to implant contact was significantly greater in non-submerged implants. These differences suggest potential advantages of the submerged procedure in the long-term.

Additionally, the authors argue that the maxillary (top jaw) incisor teeth as used in this experiment could become a preferred model instead of the more commonly used mandibular (lower jaw) teeth, due to the similarity of greyhound maxillary teeth to human incisors. These similarities, they argue, are of particular importance since ‘most of the immediate implant studies’ aim to asses ‘aesthetic outcomes’ (p558,560) for humans. Because of this consideration, and the potential differences between submerged and non-submerged procedures the authors conclude this topic ‘warrants further investigation’.

The authors also noted that there were wide standard deviations of data within each animal on some variables, and that the differences did not reach statistical significance.


The authors of this study mention ‘aesthetic’ concerns at several points – “Last but not least, considering that most of the immediate implant studies are aiming to assess factors that could affect the aesthetic outcomes in the anterior maxilla” (p560). Given that vanity is one of the primary considerations in this and similar studies, this makes clear the importance of concern for the pain and loss of life of the animals used in such experiments.

Several similar studies have already been undertaken with humans which use x-ray for analysis, rather than the bone and tissue analysis (histomorphometrics) “designed to compare the effect of sub-merged and NS healing on hard tissue alterations at the facial surfaces of immediate implants” as used in this experiment. An absence of bone and tissue studies is cited as the primary impetus for this experiment. However, there was an earlier study with a bone and tissue analysis on submerged and non-submerged implants published in 2010 and not cited by the authors (1). Given this, and the number of other similar studies which were cited by the authors, was this experiment truly treading new ground and worthwhile?


The experiment was funded by a grant from the ITI Foundation for the Promotion of Implantology, Switzerland.

What You Can Do

Click here to sign our Petition demanding an end to the use of dogs in experiments.


(1) Caneva, M., Botticelli, D., Salata, L, A., Scombatti Souza, S, L., Bressa, E., Lang, N, P. (2010) ‘Flap vs. “flapless” surgical approach at immediate implants: a histmorphometric study in dogs’ in Clinical Oral Implants Research, Dec, 2010, Vol.21(12), p.1314-1319.


Stay informed

Sign up to our newsletter for all the latest news and SMS campaign alerts.


Add a donation to help HRA carry out our vital work.
See what we can do with your contribution in 2023.