In our study,
we identified that the ADC calculations obtained from diffusion weighted images at different b levels is highly sensitive and specific in the differentiation of soft tissue sarcoma recurrence and postoperative benign changes.
Conventional MRI has an important role in the postoperative evaluation of soft tissue sarcomas.
However,
once patients are delivered radiotherapy and chemotherapy after surgical treatment,
differentiating areas of posttreatment tissue changes from tumor recurrence can sometimes be difficult (2).
Postoperative seromas,
hematomas and inflammatory changes following radiotherapy are hyperintense in T2 weighted sequences and most tumor recurrences also appear hyperintense in T2 weighted sequences (912).
Vanel and colleagues have shown that hypointense signal changes in T2 weighted images have a sensitivity of 96% in ruling out tumor recurrence and residues (11).
In addition,
the preservation of the muscle texture sign finding in T1 weighted images can also be used to rule out tumors (13).
In dynamic evaluations,
the areas of posttreatment tissue changes enhance with contrast in the later stage and tumor recurrence enhances in earlier stages (14).
The sensitivity of dynamic evaluation has been reported as 87% and the positive predictive value as 70% (15).
Myxoid tumors can be confused with postoperative seromas because of their high signal intensities in T2 weighted images (16,17).
Tumors that contain high mineralization and fibrotic components may be confused with postoperative scar tissue (2).
Tumors with hemorrhagic components may be confused with postoperative hematomas.
Therefore,
the tumor recurrence can be differentiated from posttreatment tissue changes to a limited extent in some tumors (10).
In cases where conventional methods fall short and if contrast is not administered to the patient,
new imaging methods are needed to differentiate recurrent tumors from postoperative tissue changes and to guide the surgery.
ADC mapping in diffusion MRI evaluations is used to demonstrate the levels of cellularity in different areas of the body.
Diffusion weighted images can be used to differentiate tumor recurrence from posttreatment soft tissue changes.
Low ADC values are obtained in malign tumors with high cellularity,
fibromuscular tissue and fat tissue; high ADC levels are obtained from tumors with low cellularity,
lesions with fluid contents,
necrotic regions and acellular regions (7,9).
Most of our patients included in the study had highgrade sarcomas that demonstrated high cellularity.
However,
we could not identify a meaningful correlation between the histopathological grades and the ADC values.
The ADC values of relapsed massive lesions were lower when compared to the areas of benign changes.
When the ADC values of benign changes like hematomas,
seromas,
edema and inflammatory changes that develop after treatment were compared with recurrent tumors,
they were higher and these results are consistent with literature.
Low ADC values are obtained in the development of fibrotic tissue,
another benign change (18).
In the study conducted by Del Grande and colleagues,
it was shown that lower ADC values were obtained from fibrotic tissue areas than the tumor tissues.
This is why we did not perform calculations in the areas we regarded as fibrotic tissue that had hypointense signal characteristics in T1T2 weighted images and did not enhance with contrast after IV contrast administration in our study.
Baur and colleagues have reported that ADC mapping can be used to identify sarcoma recurrence and that diffusion ADC mapping is effective in differentiating muscular edema and postoperative seromas from recurrent tumors (19).
In the study conducted by Grande and colleagues with a little number of patients using 3 tesla MRI,
they found that the ADC values obtained from diffusion weighted MRIs were statistically significantly different for recurrence,
postoperative scar tissue and hematomas.
They had identified that the mean ADC for recurrence was 1.08 x103 mm2/sec,
0.9 x103 mm2/sec for postoperative scar tissue and 0.9 x103 mm2/sec for hematomas (18).
We calculated the mean ADC value of relapsed masses as 1.69 x103 mm²/sec and as 2.98 x103 mm²/sec for areas of posttreatment tissue changes.
Because we did not measure the ADC values of fibrotic tissues in our study,
we may have calculated a higher ADC value for posttreatment changes than the ADC value calculated by Grande and colleagues for postoperative scar tissue.
When the tumors in our study were evaluated according to the histological sub types,
it was seen that 2 angiosarcomas had higher ADC values than the other soft tissue sarcoma recurrences (mean 2,7 x103 mm²/sec ).
The ADCs obtained might be higher because of the different histological contents of angiosarcomas like hemorrhages.
The role of different b values in showing soft tissue sarcomas and sarcoma recurrences has not been investigated before (18,20).
Different b values can change the sensitivity of the diffusion.
At lower b values,
a signal decrease may occur in water molecules that have high levels of motion similar to intravascular space.
This is why ADC is significantly affected by vascular perfusion.
At high b values the signaltonoise ratio decreases and this also affects ADC calculations (1924).
Due to this,
to obtain accurate ADC values,
calculations must be done with at least 3 different b values (20).
We did 4 different ADC calculations in our study.
The mean ADC calculations at b values of 50 (2,10 x103 ±0,57,
P<0,01) were higher when compared to other b values but we found meaningful results in showing recurrence just as in other b values.
We found that all b values had high sensitivity and specificity and identified that the sensitivity and specificity was highest at b 1000 values.
To avoid artefacts it may be sufficient to shorten the duration and carry out evaluations at 2 different b values.
Because the signaltonoise ratio is lower in diffusion weighted images,
more artefacts may be present especially in the postoperative period when compared to conventional sequences (1920).
We did not include patients who had prostheses or operation material that cause artefacts in the study.
Our study had some limitations.
Our study is retrospective and we worked with a heterogeneous patient group.
Because the tumors were located in different areas of the body the imaging protocols differed accordingly.
Because we used 5 different b values the imaging duration was prolonged and movement artefacts were more common.
Another limitation was number of the patients.
Both the overall number of the patients and relapse patients numbers were low Our patient and relapse patient numbers were low (15/40). Further research,
with more patients directed to the ADC evaluation of specific sarcoma diagnoses and research that evaluates the ADC calculations of benign neoplasias like lowgrade tumors and fibromatosis that have high recurrence rates are necessary.
Conclusion:
The ADC values obtained from diffusion weighted images have high sensitivity and specificity in differentiating recurring soft tissue sarcomas during monitoring after treatment from postoperative changes.
The highest sensitivity and specificity was obtained at the b 1000 level.