Long-Term Outcomes and Prognostic Factors Affecting Survival after Pulmonary Metastasectomy in Solid Tumors of Childhood: A Single Center Experience.

Introduction: This study assessed the long-term survival and the prognostic variables affecting survival following pulmonary metastasectomy (PM) secondary to childhood solid tumors. Materials and
Methods: A retrospective analysis was done on 22 children who underwent PM for solid tumors between January 2007 and February 2020. The overall survival (OS) and event-free survival (EFS) at the end of the study period were noted. Tumor histology, completeness of resection, disease-free interval, laterality, location, number, and size of lung nodules were assessed for their significance in contributing to survival.
Results: High-grade osteosarcoma (54.5%), followed by Wilms' tumor (18.2%), was the most common histological types. Unilateral nodules (59.1%) situated in a peripheral, sub-pleural location (91%) were the most common presentation. Pleural extension was noted in 12 (54.5%) patients. Synchronous pulmonary metastases were noted in 12 (54.5%) patients. Two developed metastases while undergoing chemotherapy and eight after the completion of therapy. The EFS and OS were both 31.8% at a median follow-up of 15.5 months (range 3-129 months). The median time required for an event to occur was 4 months (95% confidence interval [CI]: 1.4, 6.6 months) and median post-PM survival interval was 17 months (95% CI: 6.6, 27.4 months). Significant association was noted between preoperative tumor response to chemotherapy (P = 0.002) and survival.
Conclusion: PM can improve survival in a select group of children with metastatic solid tumors. Favorable tumor response to chemotherapy was found to be a significant prognostic factors influencing survival. Copyright:

INTRODUCTION

The management and prognosis of metastatic solid tumors in the pediatric and adolescent population have significantly evolved from palliation to a highly specialized multidisciplinary approach, often with a curative intent. This has been made possible by improved chemotherapeutic agents, widening surgical armamentarium, enhanced imaging, and more focused and rationalized radiation therapy.

It is modern surgical dogma that pulmonary metastasectomy (PM) be undertaken after complete clearance of the primary tumor, confirming the absence of extra-pulmonary metastatic involvement, and ensuring sufficient residual pulmonary reserve.[1] This study seeks to assess the long-term survival and the prognostic variables affecting survival following PM secondary to solid tumors in children.

MATERIALS AND METHODS

All children and adolescents at or below 16 years at the diagnosis of primary tumor who underwent treatment in the pediatric hematology oncology unit for solid tumors with isolated lung metastases and underwent PM in the department of pediatric surgery between January 2007 and February 2020 were included in the study. Institutional Review Board approval was obtained before the study.

Patients were divided into those with pulmonary metastases at the diagnosis and those that occurred at relapse. For those in whom pulmonary metastases occurred at recurrence, disease-free interval (DFI) was defined as the period from resection of the primary tumor until detection of metastatic pulmonary disease. Response to chemotherapy was defined as “good,” if there was radiological evidence of either decrease in the size or number of nodules, or both. The converse was defined as “poor.” Completeness of resection was defined as complete clearance of all tumors, including any pleural extensions, as evidenced by computed tomography positron emission tomography (CT/PET) CT. The outcomes assessed were the post-PM overall survival (OS), event-free survival (EFS), and the prognostic factors affecting survival. OS was defined as the period from PM to the last follow-up/death of the patient. An “event” was defined as recurrence, progression, or death.

Statistical analysis using the median and range was used for quantitative skewed data and number and percentage for categorical data. Nonparametric Kruskal–Wallis test was used to assess the association between the mode of access (open versus thoracoscopic) and postoperative hospitalization. Nonparametric Mann–Whitney test was used to compare the outcomes (two groups) on skewed quantitative variables. The Fisher's exact test (where cell count was small) and Yate's continuity correction (where the cell value was zero) were used to find an association between the categorical variables. Kaplan–Meier analysis was performed to find the median post-PM OS and EFS. Intraclass correlation coefficient (ICC) was used to assess the agreement between the number of nodules detected by CT with that assessed at surgery, as well as, between the size of the largest nodule detected by CT with size of the largest nodule confirmed by histopathological examination. All tests were two-sided at α = 0.05 level of significance. All analyses were done using the Statistical Package for the Social Sciences (SPSS) software version 21.0 (Armonk, NY, USA: IBM Corp).

RESULTS

A total of 22 patients were included in the analyses. The demographic and baseline characteristics of the patients are depicted in Table 1. The median age at the diagnosis of primary was 10 years (range 1–16), with a male predominance (77.3%). High grade osteosarcoma (OSA) (54.5%), followed by Wilms’ tumor (WT) (18.2%), was the most common histological types encountered. Unilateral nodules (59.1%) situated in a peripheral, sub-pleural location (91%) were the most common presentation. Pleural extension was noted in 12 (54.5%) patients.

Table 1

Demographic and baseline study characteristics (n=22)

Parameter	Median (minimum, maximum)/n (%)
Age (years)	10 (1, 16)
Gender
Male	17 (77.3)
Female	5 (22.7)
Histology of primary
Osteosarcoma	12 (54.5)
Wilms’ tumor	4 (18.2)
Alveolar rhabdomyosarcoma	2 (9.1)
Paraspinal ganglioneuroma	1 (4.5)
Ewing sarcoma	1 (4.5)
Adrenocortical carcinoma	1 (4.5)
Hepatoblastoma	1 (4.5)
Incidence of pulmonary metastases relative to treatment of primary
Preceding initial treatment	12 (54.5)
While on treatment	2 (9.1)
Succeeding treatment	8 (36.4)
DFI for metachronous lung metastases (months)	20.5 (5, 30)
Laterality
Unilateral	13 (59.1)
Bilateral	9 (40.9)
Location
Peripheral	20 (90.9)
Central	1 (4.5)
Both	1 (4.5)
Pleural extension
No	12 (54.5)
Yes	10 (45.5)
Neoadjuvant chemotherapy before PM
Yes	16 (72.7)
No	6 (27.3)
Preoperative tumor response to chemotherapy
Poor	13 (81.2)
Good	3 (18.8)
Not applicable-upfront surgery	6
Tumor response to chemotherapy based on histology	Good	Poor	NA
Osteosarcoma	0	9 (75)	3 (25)
Wilms’ tumor	2 (50)	1 (25)	1 (25)
Alveolar rhabdomyosarcoma	0	2 (100)	0
Paraspinal ganglioneuroma	1 (100)	0	0
Ewing sarcoma	0	1 (100)	0
Adrenocortical carcinoma	0	0	1 (100)
Hepatoblastoma	0	0	1 (100)
Parameter	Median (minimum, maximum)/n (%)
Mode of surgical access
Open thoracotomy	18 (81.8)
Thoracoscopic	2 (9.1)
Combined	2 (9.1)
Type of excision
Wedge	20 (90.9)
Lobectomy	2 (9.1)
Number of nodules on CT	2 (1, 17)
Number of nodules excised	2 (1, 16)
Largest nodule on CT (cm)	1.6 (0.75, 7.6)
Largest nodule on histopathology (cm)	2.0 (0.5, 8.5)
Completeness of resection
No	2 (9.1)
Yes	20 (90.9)
Number of revision PM surgeries for recurrence
0	20 (90.9)
1	1 (4.5)
2	1 (4.5)
Postoperative hospital stay (days)	3.5 (1, 6)
Immediate postoperative complications
No	20 (90.9)
Yes	2 (9.1)
Post-PM follow-up (months)	15.5 (1, 124)
Final outcome
Alive and well	7 (31.8)
Expired	15 (68.2)
Post-PM EFS	4.5 (1, 129)
Post-PM OS	15.5 (3, 129)

PM: Pulmonary metastasectomy, CT: Computed tomographic, EFS: Event-free survival, OS: Overall survival

Pulmonary metastases were noted at the diagnosis in 12 (54.5%) patients, two developed metastases while undergoing chemotherapy, and eight after completion of therapy. All patients, except two with metachronous pulmonary metastases (one each with Ewing sarcoma [ES] and OSA) received the standard metastatic chemotherapy protocol, specific for that particular tumor. These two patients were provided with experimental therapy. While the OSA patient had a favorable outcome, the ES patient succumbed to his illness. Sixteen patients received chemotherapy before PM and six patients underwent “up-front” PM. Among those 16 patients that received chemotherapy before PM, only three patients demonstrated a favorable response which was confirmed radiologically, with diagnoses of WT (n = 2) and neuroblastoma (NB) (n = 1).

Pulmonary metastases were detected using either CT (n = 18) or PET CT (n = 4). Imaging revealed a median of 2 (range 1–17) metastatic nodules with the median size of the largest nodule being 1.6 cm (range 0.75–7.6 cm). The median interval between the surgery for the primary lesion and PM was 11.5 months (range 0–30) and the median interval from the detection of pulmonary metastases to PM was 7 months (range 0–23). At surgery, a median number of 2 (range 1–16) nodules were detected with the median size of the largest nodule on histopathological examination being 2.0 cm (range 0.5–8.5 cm). There was agreement between both the number and size of the nodules detected by imaging and those detected at surgery (ICC = 0.99 and 0.94, respectively), including the OSA histological sub-group. Of the 22 patients, there were 31 unilateral thoracic surgeries performed (26 unilateral thoracotomies and five unilateral thoracoscopies) primarily, of which 30 were wedge resections and one lobectomy (13 unilateral and nine bilateral).

Completeness of resection was achieved in 95.5% (n = 21) of patients. Our patient with ES of the left proximal humerus had metachronous pulmonary metastases. Surgery revealed posterior parietal involvement with adherence to the arch of the aorta. Incomplete resection and poor response to adjuvant therapy resulted in a poor outcome. Two other patients with OSA had pulmonary disease recurrence and underwent repeat PM. The first underwent a right thoracoscopic wedge resection for a right lower lobe nodule. Two months later, he developed bilateral lung lesions for which he underwent bilateral synchronous thoracoscopic excision. Thoracoscopy was attempted due to the subpleural location with few foci of metastases on CT, at the discretion of the treating surgeon. This was followed by bilateral lung irradiation and adjuvant chemotherapy. He then relapsed 8 months later and underwent a right thoracotomy; however, he further progressed and expired 4 months later. The second child initially underwent bilateral synchronous open thoracotomy with complete resection. He had further disease recurrence which necessitated a right lower lobectomy 3 months later. However, he had a second recurrence postoperatively and expired 5 months later.

The median postsurgery hospital stay was 3.5 days (range 1–6), with discharge following removal of the intercostal drain in all patients. Two patients developed transient complications during the immediate postoperative period: One patient with a metastatic OSA sustained a transient bronchopleural fistula which resolved in 6 days after drainage. Another patient with ES developed postoperative pyrexia due to pneumonitis which resolved with a course of intravenous antibiotics.

The EFS and OS were both 31.8% at a median follow-up of 15.5 months (range 3–129 months). The median time required for an event to occur was 4 months (95% confidence interval [CI]: 1.4, 6.6 months), and median post-PM survival interval was 17 months (95% CI: 6.6, 27.4 months) [Figure 1].

Figure 1

Mean postpulmonary metastasectomy survival

The association of various prognostic factors affecting survival outcomes, namely tumor histology, completeness of resection, DFI, laterality, location, number, and size of lung nodules were assessed [Table 2]. Significant association was noted between preoperative tumor response to chemotherapy (P = 0.002) and survival. None of the other prognostic factors had a statistically significant association with survival.

Table 2

Association between prognostic factors and outcome (n=22)

Prognostic factor	Outcome		P
	Alive (n=7), n (%)	Expired (n=15), n (%)
Age (years)*	7 (5, 11)	12 (4, 14)	0.77
Gender#
Male	5 (71.4)	12 (80.0)	1.0
Female	2 (28.6)	3 (20.0)
Histology of primary#
Osteosarcoma	3 (42.9)	9 (60.0)	0.22
Wilms’ tumor	3 (42.9)	1 (6.7)
Alveolar rhabdomyosarcoma	0	2 (13.3)
Paraspinal ganglioneuroma	1 (14.3)	0
Ewing sarcoma	0	1 (6.7)
Adrenocortical carcinoma	0	1 (6.7)
Hepatoblastoma	0	1 (6.7)
Incidence of pulmonary metastases relative to treatment of primary#
Preceding initial treatment	3 (42.9)	9 (60)	0.30
While on treatment	0	2 (13.3)
Succeeding treatment	4 (57.1)	4 (26.7)
DFI for metachronous pulmonary metastases*	28 (21, 30)	15 (5, 23)	0.053
Laterality#
Unilateral	4 (57.1)	9 (60.0)	1.0
Bilateral	3 (42.9)	6 (40.0)
Location#
Peripheral	6 (85.7)	14 (93.3)	0.27
Central	1 (14.3)	0
Both	0	1 (6.7)
Pleural extension#
No	5 (71.4)	8 (53.3)	0.648
Yes	2 (28.6)	7 (46.7)
Neoadjuvant chemotherapy before PM#
Yes	3 (42.9)	13 (86.7)	0.05
No	4 (57.1)	2 (13.3)
Preoperative tumor response to chemotherapy#
Poor	0	13 (100)	0.002
Good	3 (100)	0
Not applicable (n=6)	4	2
Interval between diagnosis of pulmonary metastases and PM (months)*	3 (1, 20)	7 (5, 12)	0.59
Mode of surgical access#
Open thoracotomy	7 (100)	11 (73.3)	0.32
Thoracoscopic	0	2 (13.3)
Combined	0	2 (13.3)
Type of excision#
Wedge	6 (85.7)	14 (93.3)	0.27
Lobectomy	1 (14.3)	1 (6.7)
Number of nodules on CT*	2 (1, 3)	2 (2, 6)	0.28
Number of nodules excised*	2 (1, 3)	2 (1, 6)	0.72
Largest nodule on CT (cm)*	2 (1.6, 3.5)	1.5 (0.9, 2.8)	0.15
Largest nodule on histopathology (cm)*	2 (1.7, 5.2)	2 (1.3, 3.0)	0.39
Completeness of resection#
No	0	1 (7.1)	1.0
Yes	7 (100)	13 (92.9)
Number of revision PM surgeries for recurrence#
0	7 (100)	13 (86.7)	0.59
1	0	1 (6.7)
2	0	1 (6.7)
Postoperative hospital stay (days)*	4 (3, 5)	3 (2, 4)	0.19
Immediate postoperative complications#
No	7 (100)	13 (86.7)	0.59
Yes	0	2 (13.3)

*Values are presented as median (minimum and maximum) and P value was obtained from nonparametric Mann-Whitney U-test, #Values are presented as n (%) and P value was obtained from Fisher’s exact test (where cell count was small) and Yate’s continuity correction (where the cell value was zero). DFI: Disease-free interval, PM: Pulmonary metastasectomy, CT: Computed tomography

DISCUSSION

Historically, the outcome of childhood solid tumors with metastases confined to the lung has been poor. A paradigm shift in philosophy was spearheaded by Richardson, who advocated PM with a curative intent in children.[2] At present, aggressive surgical management of lung metastases in children has enhanced the overall 5-year survival following PM to 20%–40%.[3] The aim of PM should be complete tumor resection with curative intent using either parenchyma-sparing wedge or precision resection. This ought to be attempted only if complete clearance of the primary tumor can be achieved, all the pulmonary metastatic lesions can be completely excised, there is no hilar involvement, the absence of extrapulmonary metastatic involvement has been confirmed, and sufficient residual pulmonary reserve post-PM is ensured.[1] Anatomical resections such as segmentectomies, lobectomies, and pneumonectomies for more centrally located lesions should be avoided to help preserve lung function.[4]

Various studies have elucidated the different prognostic factors influencing the survival: Histology of the primary tumor and its response to adjuvant therapy significantly influences outcome.[5] Depending on the response to systemic and local adjuvant therapy, primary tumors are divided into two categories: “Good responders” were WT, rhabdomyosarcoma (RMS), hepatoblastoma (HB), ES, and NB. “Poor responders” included OSA, adrenocortical carcinoma (ACC), and non-RMS soft-tissue sarcoma. In poor responders, surgery is the primary modality of treatment for both local and metastatic sites. Irrespective of the status of lung disease, good responders who achieve complete clearance of the primary tumor have better outcomes.[6] However, a subset of “good responders” such as alveolar RMS, ES, and HB has a high propensity for local tumor recurrence which are usually resistant to subsequent adjuvant therapy. Therefore, the indications for PM in different histological tumor types vary, with diverse outcomes.

WT with lung metastases comprises about 10% of cases at initial diagnosis. Survival in these patients remains high (up to 83%).[7] PM is rarely warranted in metastatic WT, except in cases where response to chemotherapy is suboptimal. PM of residual pulmonary metastases may help avoid lung irradiation (SIOP protocol), and its subsequent complications such as pulmonary fibrosis and breast cancer. The SIOP study comprising 234 patients of WT with pulmonary metastases reported only 17% requiring PM with an OS of 92%.[8] We had four patients with WT in our study who underwent PM, with a favorable outcome in 75% of the cases; this was comparable to the reported 5-year survival of 80%.[5] One child who was noncompliant with follow-up visits presented a year post-PM with inoperable local tumor recurrence.

Nearly 18% of RMS present with isolated lung metastases.[9] This cohort had relatively better survival when compared to extrapulmonary metastases (42% vs. 24%, 3-year EFS). Considering the subtypes of RMS, embryonal variant have isolated lung metastasis in 3% cases, with a 5-year EFS of 37%–41% and an OS of 49%–52%.[1011] Although primary management in the embryonal variant with isolated lung metastases is systemic therapy, PM has been advocated in recurrent metastases, resistance to adjuvant therapy, those with malignant pleural effusions, and in older patients.[11] The alveolar variant of RMS with isolated lung metastases has a poorer prognosis, with EFS less than half when compared to other histological variants (18% vs. 37%).[9] This is due to its high propensity for local recurrence and subsequent resistance to therapy. PM in these cases may be indicated only in those with isolated stable disease.[7] Both patients with RMS in our cohort had alveolar variants and post-PM, developed recurrent inoperable tumors at the primary tumor site.

NB metastasizing to the lung is extremely rare, with an overall incidence of synchronous metastases at 0.4%–3.7%.[712] PM in metastatic NB is usually undertaken for histopathological diagnosis, increase in size of the nodule on therapy, or uptake on a functional scan. In stage IV NB, the 5-year EFS and OS were worse for those with pulmonary metastases than those with extrapulmonary spread (EFS 30% vs. 38%; OS 34.5% vs. 44.7%).[1213] In our cohort, we had one patient with metastatic paraspinal NB where the lung nodule increased in size despite clearance of the primary. Histology of the lung lesion was reported as ganglioneuroma, indicating tumor maturation.

About 10%–15% of ES present with lung metastases[514] with a reported 5-year survival of 25%–40%.[515] According to the Euro Ewings protocol, which is presently followed in our institution, chemotherapy and whole lung irradiation have been the cornerstones of therapy in these patients. Although highly sensitive to adjuvant therapy, surgical clearance is important as these tumors have a high propensity for recurrence. Studies have shown PM to improve 5-year survival in isolated metachronous disease to 58%.[16] However, ES lesions are often soft, fleshy, and blend in with the surrounding parenchyma, making complete excision difficult and compromising the therapeutic benefit of PM.[7]

HB with synchronous metastases to the lung may be present in 20% of cases,[17] with a 5-year survival of 25%–50%.[8] Uncontrolled primary disease is an absolute contraindications to perform PM.[18] PM being done prior to hepatic resection may be the preferable sequence in HB surgery, as doing a major hepatic resection is futile if complete disease clearance of the lung cannot be achieved. A hybrid approach, as our practice, is commonly employed in metastatic HB to the lungs where initial chemotherapy is backed by PM for chemo-resistant residual lung lesions.[7] Even repeat PM with adjuvant chemotherapy has been described for recurrent lung secondaries with reasonable success (80%), provided the recurrence is confined to the lung.[17] However, a recurrence in the primary site is mostly associated with a fatal outcome. We had a boy with metachronous metastatic HB of the right lobe of the liver who underwent right hepatectomy followed by PM. However, he developed local relapse in the left lobe of liver 18 months later, to which he succumbed.

In OSA, the lung is the most common site for metastatic disease. While 15%–25% of patients presents with synchronous pulmonary metastases, 40%–50% with nonmetastatic disease at presentation develop secondaries later, which spread mostly to the lung.[19] Indication for PM in metastatic OSA includes postchemotherapy residual lung lesions which are amenable to complete resection in the absence of other site metastases. There should be complete resection of the local tumor because performing a PM is futile if local clearance has not been achieved. In metastatic OSA, the overall EFS was found to be 25% and the 5-year OS 23%–38%.[31920] Further sub-analysis revealed that EFS and 5-year OS were particularly poor for those developing pulmonary metastases while on therapy (7%, 6%–8%), as compared to those with synchronous (18%, 18–34%) and metachronous (25%, 31%–52%) lung metastases, respectively.[2021] In our study, OSA patients with both synchronous lung metastases and metastases developing while on chemotherapy had dismal OS (none survived). This, when compared to an OS of 50% for those with metachronous metastases, indicates that favorable response to first-line chemotherapy was an important factor associated with improved survival. This concurs with the Cooperative OSA Study Group findings.[22]

ACC is a rare tumor (0.2%) in children with 31%–33% presenting with synchronous metastases, usually to the lungs, and has a 5-year OS of 0%–24%.[2324] PM, when performed early and completely, is the treatment of choice in metastatic ACC confined to the lung, as ACC is mostly resistant to mitotane and radiation therapy.[818] Our experience was that of a girl with left-sided synchronous metastatic ACC who had intraoperative tumor rupture during excision of the primary. PostPM, she developed relapse at the primary tumor bed with peritoneal seeding and expired. Thus, meticulous care should be taken to prevent tumor spillage as ACC, despite being well encapsulated, is gelatinous and prone for intraoperative tumor rupture (20%–43%), resulting in future inoperable recurrences.[7]

Completeness of resection is an important prognostic factor influencing long-term survival.[182025] Even for tumors with poor sensitivity to adjuvant therapy and those with repeat recurrences limited to the lung, PM with negative tumor margins enhances survival.[8] In our study, 95.5% (21/22) had complete PM. The patient with ES had tumor extending through the posterior pleura up to the arch of aorta which was detected only at surgery, highlighting the limitations of radiological imaging. There were nine patients (40.9%) with pleural disruption in the form of parietal pleural and diaphragmatic involvement. All these children, excluding those with WT, had a poor outcome, irrespective of completeness of resection. Literature from the developed world has documented synchronous pulmonary metastases in 10%–40% of childhood solid tumors.[45] Our study revealed a higher incidence of synchronous lung metastases at 63.6%, implying that the disease was more advanced at initial presentation. Ahmed et al.[20] reported the OS of synchronous lung metastasis at 34.3%, lung metastasis developing during chemotherapy at 8%, and metastases developing after completion of chemotherapy at 52%. Rasalkar et al.[26] also observed that synchronous pulmonary metastases were associated with a worse prognosis. Similar findings were noted in our study, where OS was 25% for synchronous metastases, none for metastases developing during chemotherapy, and 50% for metachronous metastases. One of the children who developed metastases while on chemotherapy had favorable histology WT, implying that PM may not be beneficial for those who develop pulmonary metastases while on chemotherapy.

DFI has been described as a prognostic factor. Tronc et al[1] observed that the 5-year survival rate of a patient with DFI of at least two years was more than 49%, and it fell to 13% if the DFI was less than two years; however, Letourneau et al[27] found no significant difference. Our study revealed no statistical significance between DFI in those without lung metastases at initial presentation (n=8) and survival (p=0.053.) Larger studies are required to establish this association. We also found that, besides patients with WT and NB, all others with synchronous isolated lung metastases had a fatal outcome.

The number and size of pulmonary metastatic nodules have been adversely linked to survival[1222628]. Those with three or less nodules often fared better than those with more[21]. As compared to a single nodule, the relative risk of death doubled with four metastatic lung nodules and quadrupled with nine or more[1]. The size of the largest metastatic nodule has also been reported to have a prognostic bearing in adults[29]. However, similar to studies by Harting et al[3] and Stanelle et al[30], we found no such association.

While several studies have commented on bilateral pulmonary metastases being associated with a worse outcome than unilateral ones[12231], differences of opinion exist[330]. Our study revealed laterality to not be significant as a prognostic determinant. With respect to location of the tumour being a prognostic factor, studies have reported a 3-year survival of only 7% for central lesions as compared to 41% for peripheral lesions. This difference was surmised to be due to greater proclivity for extra-pulmonary metastases in patients with central lesions[27]. Our study found no such association, possibly due to fewer number of isolated hilar lesions.

Accurately identifying small suspicious nodules and safely accessing deeper metastatic nodules can often be challenging. Smaller nodules that are subterranean may be approached thoracoscopically, using either isolated or combined CT-guided needle/hook wire localization and methylene blue staining.[32] In our experience, thoracoscopy offered the dual benefit of lesser morbidity and access to multiple interventions with minimal adhesions. However, as tactile perception was absent, only superficial lesions or radiologically localized lesions could be accurately identified and resected. This should especially be factored in while treating OSA metastases, where nearly 50% of radiologically detected lung lesions lie deeper than 5 mm from the pleural surface and ES metastases where the lesion blends with the surrounding parenchyma, resulting in sub-optimal tumor clearance.[725] Furthermore, distinguishing a sinister lesion from postoperative scarring may be difficult by thoracoscopy alone; ultrasound guidance may help mitigate this to a certain extent.[33] We also noted no significant advantage in reduction of postoperative hospital stay, as the duration for which chest drainage was instituted was similar for both open and thoracoscopic approaches. Open thoracotomy negates all the above-mentioned disadvantages of thoracoscopy by providing tactile perception.[25] However, thoracotomy increases postoperative morbidity and causes scarring. This may hinder access in repeat surgeries and impede complete lung expansion.

Recurrences of pulmonary lesions post-PM occurring after 1 year have a better 5-year survival (55.6%) as compared to earlier recurrences (12.3%).[3] This has been corroborated by other studies.[223435] In our cohort, recurrences occurred at a median period of 3 months post-PM. However, the longest period to recurrence was 35 months post-PM, indicating the need for long-term follow-up.

Recent advances include the use of preoperative CT-guided radiotracer labeling using technetium macro-aggregated albumin.[36] This can accurately delineate nodules <5 mm in size. Accuracy is ensured by matching the excised nodule with the nodule detected on CT. Another advancement has been in the use of indocyanine green fluorescence imaging navigation with an infrared generator and camera for detecting lung metastases as small as 0.062 mm that are secondary to HB.[37] Although these may become the norm in future, presently, logistics has been the deterrent in these modalities gaining widespread acceptance.

This is one of the first studies from a developing nation that prognosticates the long-term outcomes following PM. In our experience, a higher proportion of children presented to our institution with advanced disease. This may be attributed to lack of knowledge, poor access to health-care services, or financial constraints which are common occurrences in resource-constrained nations. There were also logistical roadblocks in optimally following up the patients, as most of them came from distant locations. The limitations of our study include its retrospective design and small sample size. Due to the rarity of PM being performed, a larger multi-centric study with longer post-PM follow-up is recommended for obtaining a more definitive conclusion.

CONCLUSION

PM offers long-term survival in a select group of individuals. Favorable preoperative tumor response to chemotherapy is a significant prognostic factor influencing survival.

IRB No. 11532

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.