Article Figures & Data
Figures
- Fig 1.
True-negative result on CT scans.
A and C, Preoperative coronal (A) and axial (C) images show JNA.
B and D, Postoperative enhanced coronal (B) and axial (D) images illustrate a normal aspect of early postoperative CT when no RD is identified. There is no enhancement at the preoperative location of JNA. Hemostatic material mixed with blood is seen in the postoperative cavity.
- Fig 2.
True-positive result. Early postoperative enhanced helical CT reveals RD.
A and B, Coronal (A) and axial (B) images of show RD on the lateral edge of the pterygoids (arrow).
C and D, Coronal (C) and axial (D) images show RD in the pterygoid muscles (arrows).
E and F, Coronal (E) and axial (F) images show RD in the foramen lacerum (arrow).
- Fig 3.
False-positive result. Suspected RD at the base of the pterygoids.
A and C, Preoperative coronal (A) and axial (C) CT scans show JNA.
B and D, Early postoperative coronal (B) and axial (D) CT scans show persistent contrast with adhesive material (arrow) at the location of the initial JNA extensions. No RD was found after surgical revision.
- Fig 4.
False-negative result. Axial CT scans.
A and B, Early postoperative (A) and 6-month (B) images show no evidence of RD.
C, At 24 months after original surgery, scan shows recurrence of JNA at the base of the pterygoids (arrow).
Tables
Patient Location of Extension* Stage 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 1 1 1 1 1 1 1 1 1 1 1 IIIA 2 1 1 1 1 1 1 1 1 IIIB 3 1 1 1 1 1 1 IIIA 4 1 1 1 IIIA 5 1 1 1 1 1 1 1 1 IIIA 6 1 1 1 1 IIIA 7 1 1 1 1 1 1 1 IIIA 8 1 1 1 1 1 1 1 IIIA 9 1 1 1 1 1 1 1 1 1 IIIA 10 1 1 1 1 1 1 1 1 IIIA 11 1 1 1 1 1 IIIA 12 1 1 1 1 1 1 1 1 1 1 IIIB 13 1 1 1 1 1 1 IIIA 14 1 1 1 1 1 1 1 1 1 IIIA 15 1 1 1 1 1 1 IIIA 16 1 1 1 1 1 1 1 1 1 1 1 1 1 1 IIIA 17 1 1 1 1 1 1 1 IIIA 18 1 1 1 1 IIIA 19 1 1 1 1 1 1 IIIA 20 1 1 1 1 1 1 1 1 1 1 IIIB * Preoperative CT scans were examined for 18 locations of tumor extension, as follows: 1 indicates present; 2, pterygopalatine fossa; 3, inferior orbital fissure; 4, orbital apex; 5, cavernous sinus (anterior); 6, infratemporal fossa; 7, temporal fossa; 8, interpterygoid fossa; 9, paratubal region; 10, foramen ovale and foramen rotondum; 11, cavernous sinus (lateral); 12, parapharyngeal space; 13, sphenoid sinus; 14, base of pterygoids and clivus; 15, cavernous sinus (medial); 16, foramen lacerum; 17, optic canal; and 18, anterior fossa.
Patient Postoperative CT Day Result* Location of RD† 1 3 TN 2 4 TP 5, 8 3 1 FP 13 4 5 TN 5 5 TP 16 6 3 TN 7 2 TN 8 2 TN 9 2 TP 2, 14 10 1 TP 8 11 2 FN 14 12 4 TN 13 4 TN 14 2 TP 14 15 5 TN 16 1 FP 14 17 1 TN 18 1 TN 19 1 TP 6, 14 20 2 FN 14 * FN indicates false-negative; FP, false-positive; TN, true-negative; and TP, true-positive.
† RD locations are listed according to 18 locations of tumor extension; 1 is nasopharyngeal vault; 2, pterygopalatine fossa; 3, inferior orbital fissure; 4, orbital apex; 5, cavernous sinus (anterior); 6, infratemporal fossa; 7, temporal fossa; 8, interpterygoid fossa; 9, paratubal region; 10, foramen ovale and foramen rotondum; 11, cavernous sinus (lateral); 12, parapharyngeal space; 13, sphenoid sinus; 14, base of pterygoids and clivus; 15, cavernous sinus (medial); 16, foramen lacerum; 17, optic canal; and 18, anterior fossa.
Location No. of Observations Pterygopalatine fossa 1 Cavernous sinus (anterior) 1 Interpterygoid fossa 2 Base of pterygoids and clivus 3 Foramen lacerum 1 Note.—Several locations could have been involved in the same RD.
Parameter Reader 1 Reader 2 Reader 3 Reader 4 Value 95% CI Value 95% CI Value 95% CI Value 95% CI Sensitivity (%) 71 31, 99 75 34, 97 75 34, 97 75 40, 97 Specificity (%) 91 67, 98 100 72, 100 100 72, 100 83 59, 99 PPV (%) 83 50, 97 100 54, 100 100 54, 100 75 47, 99 NPV (%) 83 52, 98 83 53, 98 83 53, 98 83 52, 98 Prevalence (%) 39 17, 64 42 20, 66 42 20, 66 40 18, 67 Note.—If a reader responded “do not know” (four choices in three patients, or 80 choices [5%]), the choice was excluded in the calculation of sensitivity, specificity, PPV, and NPV, but it was taken into account for the calculation of interreader agreement.
