2.3. Surgical procedures and study groups

All surgical procedures were performed under sterile conditions and general anaesthesia induced by intravenous propofol (3–5 mg/kg; Propovet, Abbott). Prior to the surgery, the dogs were sedated with intramuscular medetomidine (20 μg/kg; Domtor, Esteve), and pain was controlled with morphine (0.4 mg/kg; Morfina Braun 2%, B. Braun Medical) in combination with intramuscular meloxicam (0.2 mg/kg; Metacam, Boehringer) during the surgery and 5 days after surgery.

At the end of the quarantine period, all dogs underwent extraction of the third and fourth premolars (P3 and P4) from a randomly selected hemimaxilla. The teeth were hemisected and carefully extracted with elevators and forceps in a flapless approach. To prevent infection, intravenous cefazolin (20 mg/kg; Kurgan, Normon, Madrid, Spain) and subcutaneous cefovecin (8 mg/kg SID; Convenia, Zoetis, Madrid, Spain) were administered intra‐operatively. No soft tissue or bone grafting procedures were performed, and the ridges healed spontaneously.

Three months after the tooth extraction, a full‐thickness mucoperiosteal flap was raised in the extraction region of the maxilla of each dog to instal two implants (for a total of 18 implants in the 9 dogs). Two tissue‐level standard implants (Straumann® AG, Basel, Switzerland), each with a 2.8‐mm high polished neck and an intra‐osseous part (3.3 mm in diameter and 8 mm in length), were placed in the hemimaxilla where the previous tooth extraction had been performed (P3 and P4). The distance between the centres of each implant was 10 mm, and the implants were submerged to be flush with the rough/polished surface border with the bone crest. Next, closure caps were connected to the implants. On the contra‐lateral hemimaxilla (tooth side), a buccal mucoperiosteal flap was raised from the first molar to the second premolar. Before suturing, each animal was randomly assigned to one of the three study groups (three dogs per study group, for six teeth and six implants per group): control, non‐keratinized tissue (NKT), and non‐keratinized tissue CTG (NKT‐CTG) (Figure 1).

The three study groups: (a) control group, teeth and implants showing the presence of keratinized tissue (KT) on both sides; (b) non‐KT [NKT] group, teeth and implants showing excised KT before suturing; (c) NKT‐connective tissue graft (CTG) group, showing no KT around the implants and teeth but with a connective tissue graft under the alveolar mucosa on both sides.

In the control group, the flaps were re‐positioned around the teeth and implants at the cemento‐enamel junction (CEJ) and implant shoulder (IS), respectively, in a non‐submerged approach. Thus, all teeth and implants presented a buccal band of KT of at least 2 mm.

In the NKT group, all KTs were removed from the buccal flap on the teeth and implant sides. The buccal alveolar mucosa was sutured around the teeth and implants in a non‐submerged manner. Therefore, there was no KT around the teeth (third and fourth premolar) or around the two implants.

In the NKT‐CTG group, the same procedure was performed as in the NKT group; however, the removed KT was de‐epithelialized, leaving a free CTG. This graft was then sutured to the alveolar mucosa flap, and both the graft and flap were sutured around the teeth and implants. Therefore, the NKT‐CTG group had their alveolar mucosa sutured around the teeth and implants, with a buccal CTG transmucosally submerged (Figure 2).

Photographs illustrating the surgical procedures for implants and teeth in the non‐keratinized tissue connective tissue graft (NKT‐CTG) group. (a) Baseline pre‐surgical characteristics on the implant side and (g) tooth side. (b) Full‐thickness flap elevation and implant installation; (h) flap elevation at the tooth side. (c) Excision of KT at the implant and (i) tooth sides. (d) De‐epithelization of the graft at the implant site and (j) tooth site. (e) Suturing of the connective tissue graft to the marginal alveolar mucosa at the implant site and (k) tooth site. (f) Suturing of the alveolar mucosa plus the connective tissue graft around the implants and (l) at the buccal contra‐lateral premolars 3 and 4. The NKT group underwent the same operations but without connective tissue grafts.

All groups had non‐submerged healing with resorbable sutures (Coated VICRYL RAPIDE™, Ethicon, US, LLC).

The dogs were allowed to heal for 3 months, during which plaque control was applied with a chlorhexidine spray 2–3 times a week. After the 3 months, the animals were sacrificed using an overdose of intravenous pentobarbital (40–60 mg/kg; Dolethal, Vetoquinol) after being sedated.