MRI after Whoops procedure: diagnostic value for residual sarcoma and predictive value for an incomplete second resection

Superficial soft-tissue sarcomas are malignant mesenchymal tumors located within the cutaneous and/or subcutaneous layers [1]. Ideally, magnetic resonance imaging (MRI) should be performed for local staging and surgical planning, followed by percutaneous (image-guided) biopsy to confirm the diagnosis of sarcoma, before embarking on any surgical procedure [1, 2]. However, in patients presenting with a superficial soft-tissue lesion, the differentiation between sarcoma and a benign entity can be challenging based on history taking and clinical examination [3, 4]. Furthermore, benign superficial soft-tissue lesions greatly outnumber superficial soft-tissue sarcomas [3, 4]. It has been estimated that only 1 out of 100 primary care consultations regarding new soft tissue masses are malignant [3, 5]. Because of these epidemiological data, many physicians simply do not include the possibility of sarcoma in their differential diagnosis [4]. Consequently, there are patients who inadvertently undergo a non-oncological resection for a presumed benign superficial soft-tissue lesion that proves to be a sarcoma upon pathological examination. An unplanned sarcoma resection (whether located in the superficial soft tissues of elsewhere) is called a Whoops procedure.

In patients who underwent a Whoops procedure, second surgery with tumor bed excision is recommended, with the goal of obtaining complete tumor removal with an appropriately wide margin of resection, ideally performed in a specialized sarcoma center [6]. Prior to the second surgery, MRI can be performed for operative planning. However, the value of MRI for the detection of residual disease remains unclear, with varying diagnostic values that were reported in the recent literature [7‐10]. In addition, it is currently unknown how often MRI upgrades a presumed superficial soft-tissue sarcoma to a deep-seated soft-tissue sarcoma (i.e., extending beyond the superficial muscle fascia). This information is important to the surgeon to determine the extent of the second resection. Furthermore, the value of MRI in predicting the technical outcome of the second surgery in terms of achieving a complete pathological (R0) resection [11] is currently unknown. When a complete tumor resection is not achieved with the second surgery, additional local treatment, either by means of radiation therapy and/or a third surgery, is usually deemed necessary. If MRI can identify patients in whom the second surgery is likely going to fail, this information may be helpful to the surgeon to plan a more extensive second resection beforehand.

The purpose of this study was therefore to determine the value of MRI for the detection and assessment of the anatomic extent of residual sarcoma after a Whoops procedure and its utility for the prediction of an incomplete second resection.

The results of this study show that MRI (gadolinium chelate-enhanced imaging) is very specific in the detection of residual sarcoma after a Whoops procedure but that its sensitivity is too poor to exclude residual disease. Although it can be speculated that close follow-up MRI may be a potential alternative to second surgery in patients without any visible tumor at MRI, the risk of developing (lung) metastatic disease during this “watch and wait” period may be considered unacceptably high. Therefore, MRI cannot be used to select patients who can be spared a second surgery. Second, of the 19 residual sarcomas, 5 were deep-seated of which only two were recognized as being deep-seated on MRI examination. Therefore, MRI cannot be reliably used to plan the appropriate extent of the second resection. The inability to accurately delineate the anatomic extent of residual sarcoma before the second surgery is also reflected by the fact that 42.1% of patients with a residual sarcoma underwent an incomplete second resection. Third, although it was hypothesized that several tumor characteristics at MRI (size, anatomic extent, margins, spiculae, and tail on the superficial fascia) would increase the risk of an incomplete resection, our results did not indicate such a relationship. The main limitation of MRI, similar to that of other currently available clinical imaging techniques such as ultrasonography, CT, and PET, is its inability to detect microscopic disease. New imaging techniques are required to improve operative planning and the success of the second surgery after a Whoops procedure. This clinical need may perhaps be met by using targeted, fluorescently labeled ligands combined with optical imaging instruments with micron resolution to guide surgical resection, as demonstrated by some preliminary studies [13‐15].

There are a few recent studies that investigated the value of MRI for the detection of residual tumor after a Whoops procedure. Patkar et al. [7] included 55 patients and reported MRI to have a sensitivity of 86.7% and a specificity of 90.9% when using “a focal mass typically of high signal on T2-weighted and short-time inversion recovery sequences” as criterion for residual disease. In a study by Gingrich et al. [8] in 64 patients, MRI was reported to achieve a sensitivity of 57.8% and a specificity of 89.5% when using “focal or discrete enhancing mass” as criterion for residual tumor. In another study by Wang et al. [9] in 20 patients, MRI was found to have a sensitivity of 36.4% and a specificity of 80.0% when using “nodular or mass-like enhancement” as criterion for residual disease. Finally, Erol et al. [10] enrolled 50 patients and reported MRI to have a sensitivity of 79.4% and a specificity of 82.4%, without mentioning any criteria for residual tumor at MRI. Unlike these previous studies [7‐10], the present study did not have any false-positive among a total of 26 MRI scans. Although Gingrich et al. [8] and Wang et al. [9] used the same MRI criterion for residual tumor as the present study (i.e., a contrast-enhancing nodular lesion or mass), they still had 2 false-positives among 64 MRI scans and 2 false-positives among 20 MRI scans, respectively. The false-positives in the studies by Gingrich et al. [8] and Wang et al. [9] may possibly be attributed to subacute hematomas with high T1 signal intensity on post-contrast sequences that were mistaken for residual tumors, as also shown in Fig. 2b in the study by Wang et al. [9]. To more accurately differentiate subacute hematoma from residual tumor, and to identify true enhancement, we propose to use subtraction gadolinium chelate-enhanced images, which are generated by digitally subtracting the pre-contrast from the post-contrast sequences (note that these pre-contrast and post-contrast sequences should be acquired with the same parameters for this purpose) (Fig. 5). Otherwise, the findings of recent studies on this topic are in line with that of the present study, in that a negative MRI does not reliably exclude the presence of residual disease. However, previous studies neither investigated the value of MRI for assessing the anatomic extent of residual tumor, nor its utility for predicting the pathological success of the second surgery.

The present study had some limitations. First, this study had a retrospective design, and the sample size was relatively small. Nevertheless, the consecutively enrolled cases are representative of the clinical patient spectrum that is seen in a specialized sarcoma center. Second, the value of advanced MRI techniques such as perfusion or diffusion-weighted imaging was not investigated. However, although some studies reported that these advanced MRI techniques may have value in the detection and localization of soft-tissue sarcomas [16‐19], they still have a limited spatial resolution and therefore cannot detect microscopic residual tumor. Third, this study did not irrefutably prove that MRI after a Whoops procedure has no influence on the success of the second surgery. This can only be investigated by a randomized trial. Moreover, a gross estimation of residual tumor location and size, when visible at MRI, can still be considered helpful to guide surgery. Therefore, we do not advocate to omit MRI before the second surgery but to optimize its use and interpretation and to be aware of its limitations in this setting. Fourth, at present, the surgeons at our institution mainly use the MRI scan to obtain a gross anatomic overview of where residual sarcoma is located and how deep it reaches. However, the resection itself is primarily based on macroscopic intraoperative findings, and not on the area of abnormal signal on MRI. This may have affected the correlation between MRI findings and the outcome of the second surgery. The use of intraoperative MRI may perhaps improve the success rate of the second surgery, which is a topic for further research.

In conclusion, gadolinium chelate-enhanced MRI is a reproducible method to rule in residual sarcoma, but it is insufficiently accurate to rule out and assess the anatomic extent or residual sarcoma after a Whoops procedure. Furthermore, MRI has no utility in predicting an incomplete second resection.