Kim JH, Yoon SH, Park SQ, Ban SP, Cho BK. Clinical features and treatment strategy of paraspinal arteriovenous shunt (PAVS): a systematic review with individual participants data meta-analysis.
EUROPEAN SPINE JOURNAL : OFFICIAL PUBLICATION OF THE EUROPEAN SPINE SOCIETY, THE EUROPEAN SPINAL DEFORMITY SOCIETY, AND THE EUROPEAN SECTION OF THE CERVICAL SPINE RESEARCH SOCIETY 2021;
30:2385-2400. [PMID:
33893553 DOI:
10.1007/s00586-021-06831-2]
[Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Received: 02/24/2021] [Accepted: 03/28/2021] [Indexed: 10/21/2022]
Abstract
BACKGROUND
Because of the rarity of the disease, paraspinal arteriovenous shunt (PAVS) is not well recognized, and therapeutic options remain controversial. To introduce a rare disease of PAVS and demonstrate its etiology, clinical features, treatment options, and outcomes, we presented a case report and conducted a systematic review and individual participants data (IPD) meta-analysis.
METHODS
Studies regarding on PAVS were integrated and IPD were obtained including patients' demographics, disease etiology, clinical and radiologic features, clinical courses and outcomes. Clinical manifestation and treatment outcomes were reviewed, and comparison analysis (cervical versus thoracolumbar) were performed. Further, logistic regression analyses were conducted to identify the poor prognostic factors (incomplete obliteration).
RESULTS
Fifty-two articles were selected, and 88 patients enrolled. General and location-specific characteristics of PAVSs were identified: '3/4 of the isolated and 1/4 of the associated etiology', 'bruit, thrill, or murmur (cervical) and weakness (thoracolumbar) as common symptoms', '40% multiple feeders', and '22% intradural venous involvement'. Endovascular treatment was usually preferred (75%). Of 88 enrolled patients, 18 patients showed incomplete obliteration (20.5%). In multivariate analysis, 'etiologies of systematic genetic dysplasia (P = 0.031) and trauma (negatively, 0.038)' were significantly associated with incomplete obliteration. The parameters of 'multiple feeders (0.066)' and 'combined approach (negatively, 0.065)' are verified only in univariate analysis.
CONCLUSION
General as well as location-specific characteristics of PAVS is successfully demonstrated. Approximately 20% of the incomplete obliteration is noted, and three potential poor prognostic factors are identified, namely, 'etiology of systematic genetic dysplasia (positive) and trauma (negative)', 'combined approach (negative), and 'multiple feeders'.
Collapse