A modern approach to multiple pulmonary resections in children with recurrent metastatic pulmonary disease
Abstract
Implications of repeated resections of pulmonary metastasis (PM) are not well documented in the modern era. Fifteen children underwent two (n = 8), three (n = 3), or four or more (n = 3) resections (total = 38 procedures), most commonly for osteosarcoma (71%). Operative approach included muscle-sparing thoracotomy (71%), non-muscle-sparing thoracotomy (18%), and video-assisted thoracoscopy (11%). Median resected nodules per procedure was four (range = 1–95). Prolonged air leaks were the most common postoperative complication (29%). Median hospital stay was 4 days, and no children were discharged with or have required oxygen. Event-free survival is 67% at median follow-up time of 54 months, with an overall survival rate of 64%. Repeat resection of PM appears to be well tolerated, without prolonged hospital stays or compromised pulmonary function.
Abbreviations
-
- IQR
-
- interquartile range
-
- PM
-
- pulmonary metastasis
1 INTRODUCTION
Pulmonary metastasis (PM) from extracranial solid tumors in children, particularly soft tissue sarcomas, tend to be associated with a poor prognosis.1-3 At first relapse, complete surgical resection of pulmonary disease is shown to be associated with extended survival, though recurrence is common.4, 5 Surgical approaches to metastasectomy have evolved with the pervasiveness of thoracoscopy; however, marked heterogeneity in practice still exists.6 Additionally, the implications of repeat resection of multiple nodules on hospital length of stay, chest tube duration, or supplemental oxygen requirement have not been well documented in recent literature. This study presents a single-institutional experience with an updated approach to surgical management of recurrent PM.
2 METHODS
The study was approved by the University of Michigan Institutional Review Board. A retrospective review was performed at our center of patients undergoing more than one pulmonary resection for metastatic disease, January 2008 to March 2024. Primary tumor diagnosis and therapy, operative details, postoperative course, and follow-up data were collected. Descriptive statistics are reported as frequencies or median with interquartile ranges (IQR). Chi-squared/Fisher's exact tests were used to compare categorical variables, and Mann–Whitney tests were used to compare continuous variables. Significance was defined at p < .05.
-
Detailed radiologic listing of all PMs serves as a roadmap for surgery.
-
Use of muscle-sparing thoracotomies (unless prior surgery did not spare the musculature), to allow for detailed bimanual palpation of the lung.
-
Bilateral thoracotomies can be staged or done at the same operation depending on the features of the case and overall treatment plan.
-
Avoid use of staplers to minimize lung excised, resulting scaring and foreign body reaction, which complicates future resections.
-
Minimal resection of lung tissue is prioritized by using a vessel sealer to dissect lung tissue.7
-
Closing pleura with absorbable suture to decrease postoperative scar tissue and foreign material, which improves postoperative imaging and future thoracotomies.
-
Large hilar lesions are removed with an enucleation technique following an anatomic dissection to avoid injury to the bronchus and hilar vessels. Lobectomies are reserved for large hilar lesions not amenable to this technique.
-
Use of intercostal nerve cryoanalgesia (six levels) and intrathecal narcotics for postoperative pain control.
-
Use of thoracoscopic approaches only for one to two very peripheral lesions using the techniques outlined above.
-
Early use of water seal and Heimlich valves to facilitate early discharge with air leaks, which can dramatically decrease length of stay.
3 RESULTS
Fourteen children (57% female) underwent two (n = 8), three (n = 3), or four plus (n = 3) resections (total = 38 procedures). Tumor diagnoses included osteosarcoma (n = 10), Ewing sarcoma (n = 2), synovial sarcoma (n = 1), and angiosarcoma (n = 1). Median age at diagnosis was 14 years (IQR: 12–20). Patients received systemic therapy (neoadjuvant 57%, adjuvant 21%, both 21%), while 29% also underwent radiotherapy. Most patients (64%) had PM after completion of treatment, with a median disease-free interval of 11.5 months (IQR: 5–17). Bilateral disease was found in nine patients, with seven undergoing staged procedures (Table 1).
|
Characteristic or outcome, median (IQR)/n (%)] |
First operation (n = 14) |
Second operation (n = 14) |
Third operation (n = 6) |
Four or more operations (n = 4) |
p-value (first vs. second) | p-value (second vs. third) | p-value (third vs. 4+) |
|---|---|---|---|---|---|---|---|
| Approach | |||||||
| Thoracotomy | 12 (86) | 12 (86) | 6 (100) | 4 (100) | >.99 | .347 | >.99 |
| Thoracoscopy | 2 (14) | 2 (14) | 0 (0) | 0 (0) | |||
| Operative time (minutes) | 240 (179–471) | 251 (193–393) | 188 (140–197) | 252 (207–282) | >.99 | .099 | .116 |
| EBL (mL) | 15 (5–50) | 10 (10–28) | 18 (10–50) | 20 (15–130) | .857 | .407 | .711 |
| Hospital LOS (days) | 4 (2–6) | 4 (3–6) | 4 (3–8) | 5 (4–9) | .682 | .795 | .347 |
| Prolonged air leak | 6 (43) | 3 (21) | 1 (16.6) | 1 (25) | .427 | >.99 | >.99 |
- Abbreviations: EBL, estimated blood loss; IQR, interquartile range; LOS, length of stay.
Muscle-sparing thoracotomy was used in 27 cases (71%), non-muscle-sparing in seven, and a thoracoscopic approach in four. Of patients undergoing thoracotomy, cryoanalgesia was used in 21 patients (62%) while intrathecal narcotics were used in the remaining 13 (38%). Overall, median number of resected nodules per procedure was four (IQR: 2.5–14.5; range: 1–95), with sizes ranging from 0.02 to 4.2 cm. Significantly more nodules were resected in patients who underwent thoracotomy (4) compared to those who underwent thoracoscopy (2) (p = .022). Median operative time was 254 minutes (IQR: 176–372) for thoracotomies and 190 minutes (IQR: 116–194) for thoracoscopic resections. Additional comparisons of clinical characteristics and outcomes between these approaches can be viewed in Table 2. Median blood loss was 12.5 mL (IQR: 10–50) and chest tube duration was 4 days (IQR: 2–5). There were no intraoperative or perioperative deaths in this cohort.
|
Characteristic or outcome, median (IQR)/N (%) |
Thoracotomy (n = 34) |
Thoracoscopy (n = 4) |
p-value |
|---|---|---|---|
| Nodules resected | 4 (3–18) | 2 (1–2.5) | .022* |
| Operative time (minutes) | 254 (176–373) | 190 (116–194) | .072 |
| Estimated blood loss (mL) | 20 (10–50) | 5 (2–10) | .006* |
| Chest tube duration (days) | 3 (2–5) | 1 (1–4) | .119 |
| Hospital length of stay (days) | 4 (3–6) | 2 (1.5–4.5) | .041* |
| New discharge opioid prescription | 20 (58.8) | 1 (25) | .198 |
- Note: Bold* signifies statistical significance p < .05.
- Abbreviation: IQR, interquartile range.
Overall postoperative complication rate was 28.9%, all of which was air leaks. These were managed successfully with Heimlich valve placement (n = 7), observation (n = 3), or reoperation (n = 1). Median air leak duration for patients managed with a Heimlich valve was 11 days (IQR: 9–17.5); in patients who underwent observation, median air leak duration was 6 days (IQR: 3–13).
Median length of stay was 4 days (IQR: 3–6), and discharge opioids were required after 52.6% of cases. Frequency of new discharge opioid prescriptions was significantly higher in patients who underwent thoracotomy (58.8%) compared to patients undergoing thoracoscopy (25%), though this difference did not reach significance (p = .198). No differences in discharge opioid requirement (61.9% vs. 57.1%) were found when comparing patients who received cryoanalgesia with those who received intrathecal narcotics (p = .778). No children were discharged with or have required supplemental oxygen throughout the entire study period. Event-free survival is 67% at median follow-up time of 54 months. Overall survival of the cohort is 64%, and presently seven (77.8%) have no evidence of disease.
4 DISCUSSION
PMs have important prognostic implications for pediatric sarcomatous histologies.4 This institutional study describes the feasibility of repeated pulmonary resections in a diverse cohort of patients, with relative limited morbidity and short hospital stays. Importantly, with a median follow-up period of over 4.5 years, no patients were noted to require supplemental oxygen despite some undergoing up to five lung resections. We hope our approach provides added insight for both patients and providers who are faced with recurrent pulmonary disease.
Aggressive pulmonary resection in adults has been well documented in recent years8-10; however, studies in children with metastatic disease, particularly within the last 15 years, are less common.11-15 For such a clinically complex population, modern surgical experiences must continue to inform practice, especially in the setting of enhanced treatment guidelines and improved systemic therapy options.16, 17 Thoracotomy was once the unequivocal gold standard approach for metastatic pulmonary disease. However, paradigm shift toward minimally invasive surgery for other pathologies in children18-20 has increased investigations of its applicability to thoracic surgical oncology,5, 21, 22 even culminating in an ongoing prospective, randomized control trial by the Children's Oncology Group.23 Each surgical approach confers various advantages and risks, though there continues to be marked variability in practice.6 This likely reflects the wide spectrum of presenting disease burden as well as the impetus to perform complete resection.
Despite heterogeneity of our cohort and small sample size, we believe our approach first reiterates the ongoing utility and perioperative safety of thoracotomy in this challenging population. The ability to perform thorough tactile evaluation of both parenchymal and pleural surfaces is invaluable, especially as the accuracy of computed tomography to identify all lesions preoperatively is limited.24-26 Hesitancy to subject children to repeated thoracotomy is historically rooted in associated morbidities such as prolonged hospital stay, pain, infection, scarring, and scoliosis.5, 27-30 While many of these risks must still be considered in the decision-making process, it is important to note the armamentarium of operative and perioperative techniques has also expanded over time.
Our integration of a muscle-sparing approach minimizes iatrogenic injury to structures that most commonly contribute to postoperative pain and hinder ability to perform pulmonary rehabilitation.31 Additionally, integrating synergistic modalities such as multilevel cryoanalgesia in tandem with continuous intrathecal infusions may have further downstream effects through decreasing discharge opioid requirements. While utility of cryoanalgesia for post-thoracotomy pain has been demonstrated in adult patients as early as 1980,32 evidence in children is mainly limited to surgical correction of chest wall deformities.33 Future studies would be beneficial in further elucidating its application for open metastasectomies.
Additionally, the examination of short- and long-term pulmonary function is a novel feature of the present series. Supplemental oxygen both at discharge and follow-up was not required by any patient, which we believe reflects a judicious yet intentional approach to nodule resection. With parenchymal preservation in mind, the use of a vessel sealer (instead of stapling devices) facilitates metastasectomy with excellent hemostasis and minimal removal of lung parenchyma.7 Most resection sites are then oversewn with absorbable polydioxanone suture that decreases staple artifact on follow-up imaging and for repeat thoracotomies increasing the chances of removal of all nodules. In summary, we have developed a lung-sparing technique for pulmonary metastasectomy that is well tolerated and can be a useful tool to potentially improve survival in these complex patients.
CONFLICT OF INTEREST STATEMENT
The authors have no conflicts of interest to disclose. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
