The vascular anatomy of the spine and its relationship to pyogenic vertebral osteomyelitis.

Since Batson ( 1940) drew attention to the paravertebral system of veins there have been a number of reports of cases of spinal osteomyelitis attributed to the spread of infection by venous route from the pelvis (Deming and Zaff 1942; Donohue 1949; Myrhe 1949; Hurwitz and Albertson 1950; Alderman and Duff 1952; Romanus 1953; Wood 1954; Mussey 1954; DeFeo 1954; Leigh. Kelly and Weens 1955; Henson and Coventry 1956; Lame 1956; Liming and Youngs 1956). A thorough review of the literature and a further contribution of five new cases has recently been made by Henriques (1958). Harris (1941) pointed out that the paravertebral system of veins had been adequately described by Bock (I 823) and by Breschet (1832), and that even earlier investigators were aware of it. It seems somewhat surprising that among the large series of cases of spinal osteomyelitis recorded before 1940-particularly those of Donati (1906), Volkmann (1915), Schwarz (1920), Wilensky (1934), Klein (1933), Kulowski (1936), Stammers (1938) and Turner ( 1938)-there have been only isolated instances of pelvic suppuration as a primary focus. Thus Kulowski, in sixty personal cases, saw a previous pelvic infection (prostatic abscess) in only one, and Stammers, in a smaller series, attributed the bony infection in one instance to a septic abortion. In an attempt to define the pathway by which organisms may reach the spine the vascular anatomy of the adult vertebrae has been investigated, and the more pertinent features are here described together with implications of the paravertebral venous plexus. The case histories of nineteen patients treated for osteomyelitis of the spine at this Centre from 1944 to 1958 are reviewed. For a comparison, the details of eighteen non-spinal cases of adult osteomyelitis are set out in table form. Finally the place of pelvic sepsis as a focus for osteomyelitis is discussed in the light of the anatomical and pathological findings.