Pulmonary wedge resection using VATS is the most commonly used method available for the surgical treatment of GGNs <1 cm in diameter. However, the probability of palpation failure in localising lesions can be as high as 63%.10 Thus, accurately localising the lesions is crucial.
We can classify the localisation of lesions into three types,traditionally.11,12 The first type is localisation with imaging modalities during thoracoscopy. This includes intraoperative ultrasonography13 and CT fluoroscopy14. The second type is preoperative localisation with an injection of dyes15, contrast media16, radionuclides17 or coloured adhesive agents18. The third type is preoperative localisation with hook wire or coil placement19,20. The most popular localisation technique is CT-guided hookwire or microcoil localisation. Most studies published in recent years have reported high success rates of localisation with coils or hookwires, with rates ranging from 0.4–42%.8,10,20,21.New methods, sunch as injecting indocyanine green (ICG) under the guidance of electromagnetic navigation bronchoscope,may cost more money and requires more equipment.22
Common adverse events of localisation and their cause
Complications of coil and hookwire are similar.23,24 Dislodgement is the most common cause for operation failure. Iwasaki reported that wire dislodgement occurred in up to 20% of cases.25 Among our occurrences of placement failure, 10 (40%) were dislodgement of marking materials, of which 7 and 3 were marked by hookwires and coils, respectively. According to Mullan et al.26, a wire is generally dislodged at one of three times: during transportation of the patient to the surgical suite, during surgical deflation of the lung, or during resection, when the surgeon often applies gentle retraction to the wire. Coil dislodgement occurs less than wire dislodgement does, because its rough fibre coating induces coagulation and increases adhesion to the lung tissue, and its tension fixes it on lung tissue cracks. Nonetheless, a wide needle passage or tiny pulmonary elastic resistance still causes dislodgement. The coil is soft and pliable and causes less damage to lung tissue than the wire does when dislodged. Gagliano et al. 27 reported a case in which a displaced coil was uncoiled, causing less tissue damage when compared with hookwire for localisation in ex vivo goat lungs. Breakaway is a special type of dislodgement. The breakaway of marking materials was observed in 2 (8%) of GGNs. The most serious breakaway we encountered occurred when the wire tip was embedded in the chest wall. According to Seo et al21, the distance between the wire tip and pleural surface can be regarded as the only independent factor for successful localisation. When the distance is <1 cm between the marking materials and pleural surface, the marking materials can break away from the lung due to the cutting force of the wire or the tension of the coil.
A distance of >2 cm from the GGN to the tip of the wire or coil also means a failure of localisation, which occurred in four cases (16%) in this study. During the operation, the location of the GGN and safe resection margin cannot always be predicted. An unskilled operator or uncooperative patient urges to this kind of failure. In our cases, shaking and shortness of breath caused by hypertension of the patient aroused this failure. In addition, deep lesions might suffer from this failure more often.
If the tip of the wire or coil is located in different lobes of the lung adjacent to the GGN or across two adjacent lobes, difficulties will be encountered during the operation, and even normal lung tissue will be removed. A distance of <1 cm between the marking materials and adjacent pulmonary fissure is an important risk factor for this kind of failure.
In contrast to dislodgement or breakaway, the entire or a portion of a hookwire or coil sometimes gets stuck in the chest walls when the GGN is in the vicinity of the pleural surface, and meanwhile, marking materials are released into the pleural cavity.
Another important reason for unsuccessful localisation is the occurrence of complications. Common complications include pneumothorax, haemorrhage, air embolism, acute pain and so forth. The incidence rate of pneumothorax may account for nearly half of all complications of this procedure.8,10 Repeated puncture, thick puncture needle and large-diameter coil are the main reasons for pneumothorax. Generally, haemorrhage can be divided into two categories: pulmonary haemorrhage (Figure 7) and haemothorax. Intervention strategies for patients with haemorrhage are not always necessary, while minimal pneumothorax and asymptomatic haemorrhage often occur during the localisation procedure. Embolisation often occurs in wires; however, the coiled configuration and fibre coating virtually eliminate the risk of embolisation.
How to avoid failure of localisation
First, we need to determine the indications for CT-guided hookwire or coil localisation. In their study designed to establish the utility of preoperative CT-guided hookwire localisation of GGNs, Ciriaco et al. 4 concluded that it may be beneficial when the GGN size is <10 mm and/or the distance from the pleural surface is >15 mm and may be advisable when the GGN size is >10 mm and/or the distance from the pleural surface is between 15 and 25 mm. A discriminant function analysis performed by Saito et al. 28 indicated that a linear function (i.e. depth = 0.836 × size − 2.811) could be used to differentiate between undetectable and detectable small peripheral GGNs and that preoperative hookwire marking for small peripheral GGNs should be considered for nodules in regions above those. Actually, we believe that CT-guided hookwire and coil localisation, and especially the latter, have wider indications. As long as the needle reaches the lesions, there will be some promise for success in localisation.
For lesions with a distance of <10 mm from the pleural surface, the tip of the wire or coil should be extended about 10 mm beyond the edge of the lesion. If delocalisation of the chest wall or adjacent lung lobe occurs, it is widely recommended that the pleural end of the coil or wire be removed from the chest wall via thoracoscopy. The tail end of the hookwire ought not to be fixed when CT-guided hookwire localisation is completed, to avoid pulling to deflation of the lung. The coil should not have a thick diameter to avoid a wide needle insertion route. These measures can contribute to reducing the incidence of dislodgement or breakaway.
When a GGN hides in the inner side of the shoulder blades and blocks the route through which the needle goes, it is advisable to place a hookwire after the upper body posture is adjusted. However, shoulder blades change their positions in the lateral position, and pulling the wire during coil localisation can avoid this failure.
The patient must be induced to remain static and relaxed. Because lesions in the lower lung are easily influenced by respiration, respiratory coordination in patients with such lesions is thus of great significance. Avoiding repeated punctures and paying more attention to pulmonary vessels and airways will assist in preventing complications.
What to do after an unsuccessful localisation
Under certain circumstances, despite an unsuccessful preoperative CT-guided hookwire localisation, relocalisation will be attempted. Two types of localisations can be tried simultaneously.29 Though multiple percutaneous puncture produced a significantly higher incidence of pneumothorax and hemorrhage,the localizations were clinically feasible and safe.30,31
Lung nodules can be localised during surgery after every effort is exerted before operation. Palpation is the easiest way to detect GGNs during VATS. Suzuki et al. 11 demonstrated that in cases of lesions of ≤10 mm in size, if the distance to the pleural surface is >5 mm, the probability of failure to detect the lesions is >50%; when the distance is >10 mm, the failure probability is 100%. However, we detected 16 (64%) lesions by palpation. Finger palpation is the simplest method. Radiographic findings on preoperative CT images puncture site and the position of the unsuccessful localisation materials in the lung all help to detect small pulmonary lesions during thoracoscopic exploration. The success rate of palpation by which lesions were detected can be verified. Based on the radiographic findings on preoperative CT images, puncture site and the position of unsuccessful marking materials in the lung, detecting the highly suspected area of lesion in lung tissue is feasible. A point we raise with regard to the detection of lesions by palpation is to clamp this area with sponge forceps and run a finger over this area along a straight line. In this way, even a slight difference in the sense of touch can be felt. However, this technique does not usually work if the lesion is located deep in the lung parenchyma.
In cases in which palpation failed to localise the nodule, Suzuki et al. 18 chose to convert to thoracotomy. Resection of the highly suspected area of lesion in the lung tissue is also commonly used. Some newly designed methods or tools for the detection of pulmonary lesions have been considered effective during thoracoscopy. Ohtaka et al. 32 described that O-arm is an intraoperative imaging device that can provide CT images and that the positional relationship between the lesion and needle marking will be determined based on these O-arm CT images. Barmin et al.33 designed a new tactile mechanoreceptor, with the help of which the surgeon can see the border between normal and high-density tissue in the inspected area. Okusanya et al.34 intravenously injected indocyanine green 24 hours before surgery and claimed that, during lung resections, intraoperative near-infrared imaging can be used to detect GGNs that are poorly visualised on CT and difficult to discriminate on finger palpation. All of these new methods are considered to be additional tools for facilitating intraoperative localisation and surgical resection of nonpalpable lung lesions. Segmentectomy or lobectomy should also be considered during surgery after every traditional effort has been exerted preoperatively.Wu et al.35 reported that three-dimensional navigation combined with anatomic segmental pulmonary resection avoids the adverse factors of puncture, and can replace puncture localization for GGN located in the central region of segmental or subsegmental lung or adjacent intersegmental veins.
We acknowledge some limitations of this retrospective study. First, most of the lesions in the cases we included were subcentimetre nodules. The smaller the lesions we needed to locate, the more difficult it was to do so. In addition, our study was not designed to compare the failure rate, cause of localisation failure and the method of amending after unsuccessful localisation and so forth between the two marking materials. Moreover, we have become aware of some potential technical biases in our study. Despite these limitations, the results of this study contribute to additional significant experience on localisation failure, and we conclude that, after an unsuccessfully preoperative procedure of CT-guided hookwire or coil localisation, a second localisation, intraoperative localisation, resection of the highly suspected area, or a segmentectomy or lobectomy can be successfully attempted using VATS for resection of GGNs.