Differences in intramuscular vascular connections of human and dog latissimus dorsi muscles

Abstract

Background. Distal ischemia and necrosis of the dog latissimus dorsi muscle flap used in experimental cardiomyoplasty have been reported. However, little information on the intramuscular vascular anatomy of the dog latissimus dorsi is available. It is unclear whether there are any anatomic factors relating to the muscle flap ischemia and necrosis, and whether the dog latissimus dorsi is a suitable experimental model. Methods. To study the intramuscular vascular territories in the dog latissimus dorsi muscle, and to compare the intramuscular vasculature of the dog with that of the human, 5 fresh dog cadavers and 7 fresh human cadavers were injected with a mixture of lead oxide, gelatin, and water (200 mL/kg) through the carotid artery. Both the dog and the human latissimus dorsi muscles and neurovascular pedicles were dissected and radiographed. The intramuscular vascular anatomy of the latissimus dorsi muscles was compared. Results. Radiographs demonstrate clearly that the pattern of latissimus dorsi intramuscular anastomoses between branches of the thoracodorsal artery and the perforators of posterior intercostal arteries in the proximal half of the muscle are different between the dog and the human. In the dog muscle, vascular connections between the thoracodorsal artery and the posterior intercostal arteries are formed by reduced-caliber choke arteries, whereas four to six true anastomoses without a change in caliber between them are found in the human muscle. The portion of the latissimus dorsi muscle supplied by the dominant thoracodorsal vascular territory was 25.9% ± 0.3% in the dog and 23.9% ± 0.5% in the human. For further comparison, an extended vascular territory in the latissimus dorsi muscle was demonstrated, including both the thoracodorsal territory and the posterior intercostal territories. The area of the extended vascular territory was 52% ± 0.5% of the total muscle. Conclusions. The dog latissimus dorsi model may not be a perfect predictor of the behavior of the human latissimus dorsi muscle flap in cardiomyoplasty.