Home > Journals > Italian Journal of Vascular and Endovascular Surgery > Past Issues > Giornale Italiano di Chirurgia Vascolare 1998 June;5(2) > Giornale Italiano di Chirurgia Vascolare 1998 June;5(2):97-110



To subscribe
Submit an article
Recommend to your librarian





Giornale Italiano di Chirurgia Vascolare 1998 June;5(2):97-110


language: English, Italian

Transesophageal echocardiography in the study of ectatic conditions of the thoracic aorta. Inter- and intraobserver variability

Salvini M., Palmieri P., Bonalumi G., Odero A.

From the Department of Surgery Vascular Surgery Clinic San Matteo Polyclinic, Pavia (Head: Prof. A. Odero) Pavia University


Objective. The ­rate at ­which tho­rac­ic and thor­a­coab­dom­i­nal aneu­rysms ­expand is ­unknown. Transesophageal ech­o­car­di­og­ra­phy (TOE) ­could ­play a fun­da­men­tal ­role in estab­lish­ing the nat­u­ral his­to­ry of tho­rac­ic aor­ta ecta­sias in ­patients not giv­en sur­gi­cal treat­ment. The ­present ­paper inves­ti­gates the accu­ra­cy of the tech­nique in pro­vid­ing a reli­able esti­mate of vari­a­tions in aor­tic diam­e­ter giv­en the ­impact of ­inter- and ­intraobserv­er var­i­abil­ity.
Methods. The ­study was con­duct­ed on 10 ­patients giv­en TOE for sus­pect­ed aneu­rysms of the tho­rac­ic or thor­a­coab­dom­i­nal aor­ta. It ­involved the record­ing of stan­dard TOE ­scans (5 MHz bipla­nar ­probe, ­free set­ting) or ­SVHS ­tape and repeat­ed ran­dom read­ings of max­i­mum diam­e­ter record­ed on the ascend­ing aor­ta, aor­tic ­arch and descend­ing aor­ta by two inde­pen­dent oper­a­tors on a ­blind ­trial ­basis.
Statistical anal­y­sis. The ­data ­thus record­ed was sub­ject­ed to: AOV sta­tis­ti­cal anal­y­sis for repeat­ed read­ings (dif­fer­enc­es ­between meas­ure­ments and ­between ­scans); cal­cu­la­tion of the vari­a­tion coef­fi­cient ­over ­three read­ings on ­each seg­ment stud­ied (intraobserv­er var­i­abil­ity); assess­ment of ­intra- and ­interobserv­er con­cor­dance (cor­re­la­tion coef­fi­cient); cal­cu­la­tion of the ­mean abso­lute dif­fer­enc­es ­between read­ings and ­between oper­a­tors togeth­er ­with con­fi­dence lim­its (var­i­abil­ity ­expressed in ­terms of diam­e­ter vari­a­tions ­revealed by inde­pen­dent read­ings).
Results. Analysis of the ­results ­showed ­that diam­e­ter esti­mates ­were read­i­ly repeat­able (­intraoper­a­tor vari­a­tion coef­fi­cient 0-17%) in all aor­tic seg­ments but ­less accu­ra­cy in assess­ments of the aor­tic ­arch. The ­intra/­interoper­a­tor cor­re­la­tion coef­fi­cient was ­close to 1 on all seg­ments (a lin­e­ar rela­tion­ship ­that was sta­tis­ti­cal­ly sig­nif­i­cant) and the con­fi­dence lim­it on the ­mean of abso­lute dif­fer­enc­es was ­also ­under 1 mm, ­despite occa­sion­al abso­lute dif­fer­enc­es in of up to 4 mm on the aor­tic ­arch ­when read­ings exceed­ed the ­known lim­its of aneu­rysms ­growth.
Conclusions. The ­paper con­cludes ­that TOE is extreme­ly accu­rate in ­terms of ­inter/­intraoper­a­tor var­i­abil­ity ­when the ­mean fig­ures on the ­entire ­study pop­u­la­tion are tak­en ­into ­account. However ­when the tech­nique is ­used to ­judge the ­degree of expan­sion in a sin­gle aneu­rysms ­there may be abso­lute diver­genc­es ­that ren­der the tech­nique an unre­li­able ­basis for clin­i­cal deci­sions, so ­that any judge­ment of aneu­rysm evo­lu­tion in indi­vid­u­al ­patients ­should be for­mu­lat­ed ­with ­extreme cau­tion.

top of page