Indication for drainage for a hematogenous iliopsoas abscess: Analysis of patients initially treated without drainage.
J Orthop Sci
; 26(6): 1130-1134, 2021 Nov.
Article
em En
| MEDLINE
| ID: mdl-33317896
BACKGROUND: This study aimed to determine the indications for drainage in extended haematogenous iliopsoas abscesses (IPAs), which include both primary and vertebral osteomyelitis-related IPAs. METHODS: Sixty-three IPA patients who were initially treated with only antibiotics and no drainage were enrolled. The success (S) group included patients who were cured without drainage, while the failure (F) group included those who required open or percutaneous drainage or died. RESULTS: Compared with patients in the S group, patients in the F group (n = 15) had a higher incidence of end-stage renal disease on hemodialysis, compromised immunity, vertebral osteomyelitis of the cervicothoracic spine, other musculoskeletal infections, and multilocular abscesses. The IPAs in the F group had larger transverse and longitudinal diameters. In receiver operating characteristic curve analyses for the diameter of IPAs, the most valuable cut-off points predicting the F group were a longitudinal diameter of 5.0 cm (sensitivity, 1.0; specificity, 0.67) and a transverse diameter of 2.3 cm (sensitivity, 0.73; specificity, 0.73). A combination of both diameter cut-offs had high specificity (sensitivity, 0.73; specificity, 0.90). CONCLUSIONS: Drainage should be applied in case of a larger abscess with transverse diameter ≥ 2.3 cm and longitudinal diameter ≥ 5.0 cm. Conversely, IPAs with longitudinal diameter <5 cm do not require drainage. Haemodialysis, compromised immunity, vertebral osteomyelitis of the cervicothoracic spine, and musculoskeletal infections are risk factors of conservative treatment failure.
Texto completo:
1
Base de dados:
MEDLINE
Assunto principal:
Abscesso do Psoas
Tipo de estudo:
Observational_studies
/
Prognostic_studies
/
Risk_factors_studies
Limite:
Humans
Idioma:
En
Ano de publicação:
2021
Tipo de documento:
Article