Abstract
OBJECTIVES
The aim of this study was to compare postoperative outcomes between biologic and synthetic reconstructions after chest wall resection in a matched cohort.
METHODS
All patients who underwent reconstruction after full-thickness chest wall resection from 2000 to 2022 were reviewed and stratified by prosthesis type (biologic or synthetic). Biologic prostheses were of biologic origin or were fully absorbable and incorporable. Integer matching was performed to reduce confounding. The study end point was surgical site complications requiring reoperation. Multivariable analysis was performed to identify associated risk factors.
RESULTS
In total, 438 patients underwent prosthetic chest wall reconstruction (unmatched: biologic, n = 49; synthetic, n = 389; matched: biologic, n = 46; synthetic, n = 46). After matching, the median (interquartile range) defect size was 83 cm2 (50–142) for the biologic group and 90 cm2 (48–146) for the synthetic group (P = 0.97). Myocutaneous flaps were used in 33% of biologic reconstructions (n = 15) and 33% of synthetic reconstructions (n = 15) in the matched cohort (P = 0.99). The incidence of surgical site complications requiring reoperation was not significantly different between biologic and synthetic reconstructions in the unmatched (3 [6%] vs 29 [7%]; P = 0.99) and matched (2 [4%] vs 4 [9%]; P = 0.68) cohorts. On the multivariable analysis, operative time [adjusted odds ratio (aOR) = 1.01, 95% confidence interval (CI), 1.00–1.01; P = 0.006] and operative blood loss (aOR = 1.00, 95% CI, 1.00–1.00]; P = 0.012) were associated with higher rates of surgical site complications requiring reoperation; microvascular free flaps (aOR = 0.03, 95% CI, 0.00–0.42; P = 0.024) were associated with lower rates.
CONCLUSIONS
The incidence of surgical site complications requiring reoperation was not significantly different between biologic and synthetic prostheses in chest wall reconstructions.
Keywords: Chest wall resection, Chest wall reconstruction, Biologic prosthesis, Synthetic prosthesis, Postoperative outcomes, Surgical site complications
The positive survival outcomes associated with complete resection of chest wall tumours have led to an increase in more extensive chest wall resections [1–3].
INTRODUCTION
The positive survival outcomes associated with complete resection of chest wall tumours have led to an increase in more extensive chest wall resections [1–3]. Moreover, owing to advances in treatment strategies and perioperative care, more complex cases (i.e. irradiation, infection, redo operations) are considered for chest wall resections [1, 2, 4]. Consequently, anatomical and functional restoration of the chest wall (i.e. biomimesis), to avoid flail segments and to preserve respiratory function, has become equally more complex [4–8].
Historically, synthetic prostheses made of materials such as methyl-methacrylate, polypropylene and polytetrafluoroethylene have been used to reconstruct chest wall defects after resection. Synthetic prostheses have been reported to have an increased risk of infection, incomplete incorporation and adherence to underlying organs [5, 7, 9]. This spurred the development of materials, such as acellular dermal matrix or poly-4-hydroxybutyrate [10–12], which have a potentially lower risk of infection and incomplete incorporation but are associated with higher costs [13–16]. Since the use of biologic prostheses at present is supported by studies that include small numbers of patients or heterogeneous cohorts [9, 15–17], the ideal prosthesis remains a matter of debate.
We aimed to compare the incidence of surgical site complications requiring reoperation between biologic and synthetic prostheses in a matched cohort of patients undergoing reconstruction after full-thickness chest wall resection.
MATERIALS AND METHODS
Ethical statement
This study was approved by our Institutional Review Board (# 18-391). The need for informed consent was waived because of the retrospective nature of the analysis and the lack of identifiable patient information. This study followed the STROBE reporting guidelines [18].
Study population and data collection
All patients who underwent full-thickness chest wall resection for any indication from 1 January 2000 to 30 April 2022 in the Thoracic Service at Memorial Sloan Kettering Cancer Center were reviewed for inclusion (n = 536). Patients who did not receive a prosthetic reconstruction (n = 98) were excluded (Supplementary Material, Fig. S1). The cohort was stratified by the type of prosthesis used (biologic or synthetic). Biologic prostheses were of biologic origin or were fully absorbable and incorporable and included acellular dermal matrix, poly-4-hydroxybutyrate and polyglactin.
Data on demographic information, comorbidities, imaging, pathology, disease characteristics, postoperative outcomes and follow-up were collected prospectively during the patient’s hospital stay and from information provided at outpatient visits at our institution. For patients who received additional treatment and/or follow-up outside of our centre, information regarding their status was obtained by telephone follow-up or outside correspondence received by caregivers. Follow-up was closed on 31 October 2022. Study-specific data (tumour location, ribs resected, type of reconstruction, defect size, reconstruction-specific complications) were collected retrospectively. To account for the time effect in our study, the cohort was divided into two periods: 2000–2010 and 2011–2022.
The primary end point was surgical site complications requiring reoperation. Surgical site complications included wound infection, wound dehiscence, myocutaneous flap necrosis, seroma or haematoma, prosthesis infection and material fracture. The secondary end points were rates of 30-day complications, pulmonary complications (Society of Thoracic Surgeons version 5.21.1) and grade ≥3 complications (Common Terminology Criteria for Adverse Events version 5). Exploratively, we aimed to identify risk factors that may be associated with surgical site complications requiring reoperation in the matched cohort, representing a high-risk patient cohort.
General indications and postoperative management
Patients requiring extensive chest wall resection are discussed by a multidisciplinary panel. The general indications for the use of biologic prostheses at our institution are critical wounds (e.g. infectious environment, radionecrosis), noncritical wounds in patients with a history of radiation or redo surgery and defects with a lack of adequate myocutaneous coverage. The final decision is made at the discretion of the thoracic surgeon and is based on intraoperative findings, such as the extent of the defect and the quality and availability of the surrounding soft tissues.
Prophylactic cefazolin (2 g) is administered at the time of anaesthetic induction and is repeated every 4 h for the duration of the operation. In patients with an allergy to penicillin, clindamycin (900 mg) is administered. Postoperatively, chest tubes are removed when the output is <300 cc. Myocutaneous flaps are monitored for skin colour, turgor and temperature. Blood flow is monitored with a pencil Doppler every hour for the first 24 h, every 2 h for the next 24 h and then every 4 h until discharge. The Cook-Swartz Doppler Probe (Cook Medical, Bloomington, IN) is used to assess blood flow in cases where an external Doppler does not obtain a signal. Subfascial drains are removed when the output is <30 cc for 2 consecutive days after ambulation. Patients are discharged with subfascial drains in situ and are followed up in the outpatient clinic if the output criteria are not met before discharge [19].
Statistical analysis
Categorical variables are presented as number and percentage; continuous variables are presented as median and interquartile range (IQR). Differences in patient characteristics among the two unmatched groups (biologic and synthetic) were evaluated using chi-squared tests for categorical variables and one-way analysis of variance or Mann–Whitney U tests for continuous variables. As treatment allocation could be subject to selection bias, which may not be completely accounted for in multivariable analysis, we used integer modelling to perform pairwise 1:1 matching with a maximally tolerated standardized mean difference of 0.05 and to optimize agreement between the two treatment groups. This resulted in surgical groups with similar probabilities of undergoing each type of reconstruction, which approximates randomization [20]. The variables used for matching were age, comorbidities (pulmonary, cardiac, endocrine, others), Eastern Cooperative Oncology Group (ECOG) performance status (0, 1, ≥2), smoking status (current, former, never), previous radiation therapy, induction therapy, reoperation, previous cancer, tumour location (anterior, lateral, posterior), extended pulmonary resection, defect size, number of ribs resected and type of soft-tissue reconstruction (primary closure, pedicled flap, microvascular free flap). Patients with missing covariables were excluded from matching. All variables were balanced for 46 patients in each group, observing the greatest standardized mean difference for reoperation (−0.0486) (Supplementary Material, Table S1 and Supplementary Material, Fig. S2). After matching, differences in outcomes were evaluated using McNemar’s test for categorical variables and the Wilcoxon rank sum test for continuous variables.
Univariable logistic regression was used to assess the association between various prognostic predictors and surgical site complications requiring reoperation in the high-risk matched cohort. A multivariable logistic regression model was used to identify potential risk factors that may be associated with the primary end point, adjusted for characteristics with P < 0.25 on univariable analysis. All statistical tests were two-tailed, and P < 0.05 was considered to indicate statistical significance. The Bonferroni correction was used to correct for multiple testing. Statistical analyses were performed with R (version 4.2.1, R Foundation for Statistical Computing, Vienna, Austria), using the gtsummary, MatchIt, ggplot and cobalt packages.
RESULTS
Patient characteristics in the unmatched and matched cohorts
In total, 438 patients underwent prosthetic chest wall reconstruction and were included in the analysis. Biologic prostheses were used in 11% (n = 49) and synthetic prostheses were used in 89% (n = 389) of patients. The median follow-up was 25.2 months (range, 0.03 months to 8.9 years).
ECOG performance status was ≥1 in 20% of patients (n = 88) (biologic, 67% [n = 33]; synthetic, 14% [n = 55]; P = 0.001). Pulmonary comorbidities were present in 17% of patients (n = 74) (biologic, 29% [n = 14]; synthetic, 15% [n = 60]; P = 0.021). Other comorbidities were present in 22% of patients (n = 96) (biologic, 41% [n = 20]; synthetic, 20% [n = 76]; P = 0.001). Age, sex, smoking status, cardiac comorbidity, history of chest wall radiation, reoperation and induction therapy were comparable between groups (Table 1).
Table 1:
Patient characteristics
| Characteristic | All (n = 438) | Unmatched cohort |
Matched cohort |
||||
|---|---|---|---|---|---|---|---|
| Biologic (n = 49) | Synthetic (n = 389) | SMD | Biologic (n = 46) | Synthetic (n = 46) | SMD | ||
| Age (years) | 59 (49–68) | 60 (52–67) | 59 (48–69) | 0.08 | 60 (51–67) | 59 (48–71) | −0.04 |
| Sex | 0.15 | 0.04 | |||||
| Female | 240 (55) | 30 (61) | 210 (54) | 27 (59) | 28 (61) | ||
| Male | 198 (45) | 19 (39) | 179 (46) | 19 (41) | 18 (39) | ||
| ECOG status (n = 433) | 1.3 | 0.05 | |||||
| 0 | 345 (80) | 16 (33) | 329 (86) | 16 (35) | 17 (37) | ||
| 1 | 80 (18) | 29 (59) | 51 (13) | 27 (59) | 26 (57) | ||
| ≥2 | 8 (2) | 4 (8) | 4 (1) | 3 (7) | 3 (7) | ||
| Smoking status | 0.17 | 0.00 | |||||
| Never | 182 (42) | 20 (41) | 162 (42) | 19 (41) | 19 (41) | ||
| Former | 212 (48) | 26 (53) | 186 (48) | 25 (54) | 25 (54) | ||
| Current | 44 (10) | 3 (6) | 41 (11) | 2 (4) | 2 (4) | ||
| Pack-years (n = 433) | 4 (4–30) | 3 (0–16) | 4 (0–30) | −0.14 | 3 (0–16) | 4 (0–36) | −0.26 |
| Pulmonary comorbidity | 74 (17) | 14 (29) | 60 (15) | 0.32 | 13 (28) | 12 (26) | 0.05 |
| Cardiac comorbidity | 197 (45) | 22 (45) | 175 (45) | 0.00 | 20 (43) | 21 (46) | −0.04 |
| Diabetes | 47 (11) | 7 (14) | 40 (10) | 0.12 | 6 (13) | 6 (13) | 0.00 |
| Other comorbidity | 96 (22) | 20 (41) | 76 (20) | 0.48 | 19 (41) | 18 (39) | 0.04 |
| History of radiation | 110 (25) | 16 (33) | 94 (24) | 0.19 | 14 (30) | 13 (28) | 0.05 |
| Reoperation | 124 (28) | 13 (27) | 111 (29) | 0.04 | 13 (28) | 14 (30) | 0.05 |
| Induction therapy | 152 (35) | 19 (39) | 133 (34) | 0.10 | 18 (39) | 19 (41) | 0.04 |
Data are represented as n (%) or median (interquartile range). Percentages have been rounded and may not total 100.
ECOG: Eastern Cooperative Oncology Group; SMD: standardized mean difference.
After matching, 46 patients were included in each group (n = 92 total). ECOG performance status ≥1 (65% [n = 30] vs 63% [n = 29]; P = 0.99), pulmonary comorbidities (28% [n = 13] vs 26% [n = 12]; P = 0.81) and other comorbidities (41% [n = 19] vs 39% [n = 18]; P = 0.83) were comparable between the biologic and synthetic groups, respectively (Table 1).
Surgical characteristics in the unmatched and matched cohorts
Patients were predominately treated for primary chest wall tumours (48% [n = 211]) and recurrence or metastasis (45% [n = 199]). Additionally, 1% of patients (n = 4) underwent surgery for superimposed infection (biologic, 4% [n = 2]; synthetic, 1% [n = 2]; P = 0.30).
Extended chest wall resection with pulmonary resection was performed in 44% of patients (n = 191) (biologic, 33% [n = 16]; synthetic, 45% [n = 175]; P = 0.10). The median (IQR) operative time was 270 min (182–374) (biologic, 298 min [220–433]; synthetic, 263 min [177–358]; P = 0.013) (Table 2). Table 3 summarizes the characteristics of reconstructions. Myocutaneous (pedicled and free flap) reconstructions were used in 27% of patients (n = 118) (biologic, 37% [n = 18]; synthetic, 26% [n = 100]; P = 0.089).
Table 2:
Operative characteristics
| Characteristic | All (n = 438) | Unmatched cohort |
Matched cohort |
||||
|---|---|---|---|---|---|---|---|
| Biologic (n = 49) | Synthetic (n = 389) | SMD | Biologic (n = 46) | Synthetic (n = 46) | SMD | ||
| Indication | 0.26 | 0.19 | |||||
| Primary tumour | 211 (48) | 26 (53) | 185 (48) | 25 (54) | 23 (50) | ||
| Radiation-induced tumour | 16 (4) | 1 (2) | 15 (4) | 1 (2) | 2 (4) | ||
| Recurrence | 199 (45) | 19 (39) | 180 (46) | 18 (39) | 20 (43) | ||
| Other | 12 (3) | 3 (6) | 9 (2) | 2 (4) | 1 (2) | ||
| Tumour location | 0.36 | 0.05 | |||||
| Anterior or anterolateral | 253 (58) | 26 (53) | 227 (58) | 23 (50) | 24 (52) | ||
| Lateral | 69 (16) | 4 (8) | 65 (17) | 4 (9) | 4 (9) | ||
| Posterior or posterolateral | 116 (26) | 19 (39) | 97 (25) | 19 (41) | 18 (39) | ||
| Ribs resected, n (n = 435) | 3 (2–4) | 3 (2–4) | 3 (2–4) | −0.24 | 3 (2–4) | 3 (2–4) | −0.34 |
| Extended pulmonary resection | 191 (44) | 16 (33) | 175 (45) | 0.26 | 16 (35) | 17 (37) | 0.05 |
| Defect size (cm2) (n = 421) | 90 (52–134) | 86 (49–142) | 90 (53–133) | −0.02 | 83 (50–142) | 90 (48–146) | 0.00 |
| Operative time (min) (n = 436) | 270 (182–374) | 298 (220–433) | 263 (177–358) | 0.38 | 306 (219–464) | 307 (212–450) | 0.18 |
| Operative blood loss (ml) (n = 433) | 220 (100–500) | 300 (100–500) | 200 (100–500) | 0.01 | 300 (112–500) | 250 (78–775) | −0.14 |
Data are represented as n (%) or median (interquartile range). Percentages have been rounded and may not total 100.
SMD: standardized mean difference.
Table 3:
Reconstructive characteristics
| Reconstruction, characteristic | All (n = 438) | Unmatched cohort |
Matched cohort |
||||
|---|---|---|---|---|---|---|---|
| Biologic (n = 49) | Synthetic (n = 389) | SMD | Biologic (n = 46) | Synthetic (n = 46) | SMD | ||
| Biologic prosthesis | |||||||
| Acellular dermal matrix | 19 (4) | 19 (39) | – | 19 (41) | – | ||
| Poly-4-hydroxybutyrate | 26 (6) | 26 (53) | – | 24 (52) | – | ||
| Polyglactin | 4 (1) | 4 (8) | – | 3 (7) | – | ||
| Synthetic prosthesis | |||||||
| Methyl-methacrylate | 204 (47) | – | 204 (52) | – | 18 (39) | ||
| Polypropylene | 117 (27) | – | 117 (30) | – | 17 (37) | ||
| Polytetrafluoroethylene | 50 (11) | – | 50 (13) | – | 7 (15) | ||
| Titanium | 18 (4) | – | 18 (5) | – | 4 (9) | ||
| Soft-tissue reconstruction | 0.29 | 0.00 | |||||
| Primary closure | 320 (73) | 31 (63) | 289 (74) | 31 (67) | 31 (67) | ||
| Pedicled flap | 97 (22) | 13 (27) | 84 (22) | 10 (22) | 10 (22) | ||
| Microvascular free flap | 21 (5) | 5 (10) | 16 (4) | 5 (11) | 5 (11) | ||
Data are represented as n (%). Percentages have been rounded and may not total 100.
SMD: standardized mean difference.
After matching, extended chest wall resection with pulmonary resection was performed in 35% (n = 16) vs 37% (n = 17) of patients (P = 0.83), the median (IQR) operative time was 306 min (219–464) vs 307 min (212–450) (P = 0.42; Table 2) and myocutaneous (pedicled and free flap) reconstructions were used in 33% (n = 15) vs 33% (n = 15) of patients (P = 0.99; Table 3).
Postoperative outcomes in the unmatched and matched cohorts
Table 4 summarizes postoperative outcomes. Overall, surgical site complications requiring reoperation were observed in 7% of patients (n = 32) (biologic, 6% [n = 3]; synthetic, 7% [n = 29]; P = 0.99); 75% of these occurred within 90 days postoperatively and 94% occurred within 1 year. Prosthetic infection or material fracture was not observed in the biologic group; in the synthetic group, 2% of patients (n = 9) had prosthesis infections (P = 0.61) and 2% (n = 7) had material fractures (P = 0.99). All prosthetic infections required removal of the prosthesis, followed by vacuum-assisted closure with secondary healing (n = 2) or myocutaneous reconstruction (n = 3) in 5 patients, immediate myocutaneous reconstruction in 2 patients, primary closure in 1 patient and secondary healing in 1 patient. Overall, the 30-day complication rate was 37% (n = 162) (biologic, 43% [n = 21]; synthetic, 36% [n = 141]; P = 0.37). Pulmonary complications were observed in 13% of patients (n = 56) (biologic, 14% [n = 7]; synthetic, 13% [n = 49]; P = 0.74). Table 5 summarizes pulmonary complications. Grade ≥3 complications were observed in 13% of patients (n = 56) (biologic, 8% [n = 4]; synthetic, 13% [n = 52]; P = 0.30).
Table 4:
Postoperative outcomes
| Outcome | All (n = 438) | Unmatched cohort |
Matched cohort |
||||
|---|---|---|---|---|---|---|---|
| Biologic (n = 49) | Synthetic (n = 389) | P-Value | Biologic (n = 46) | Synthetic (n = 46) | P-Value | ||
| Resection status | 0.28 | 1 | |||||
| R0 | 381 (87) | 45 (92) | 336 (86) | 42 (91) | 41 (89) | ||
| ≥R1 | 57 (13) | 4 (8) | 53 (14) | 4 (9) | 5 (11) | ||
| Surgical site complications requiring reoperation | 32 (7) | 3 (6) | 29 (7) | 0.99 | 2 (4) | 4 (9) | 0.68 |
| Wound infection | 22 (5) | 4 (8) | 18 (5) | 0.29 | 3 (7) | 5 (11) | 0.72 |
| Dehiscence | 10 (2) | 1 (2) | 9 (2) | 0.99 | 0 (0) | 1 (2) | 1 |
| Prosthesis infection | 9 (2) | 0 (0) | 9 (2) | 0.61 | 0 (0) | 1 (2) | 1 |
| Material fracture | 7 (2) | 0 (0) | 7 (2) | 0.99 | 0 (0) | 1 (2) | 1 |
| 30-Day complications | 162 (37) | 21 (43) | 141 (36) | 0.37 | 18 (39) | 16 (35) | 0.86 |
| Pulmonary complications | 56 (13) | 7 (14) | 49 (13) | 0.74 | 5 (11) | 7 (15) | 0.77 |
| Grade ≥3 complications | 56 (13) | 4 (8) | 52 (13) | 0.30 | 2 (4) | 6 (13) | 0.29 |
| Length of stay (days) | 6 (4–8) | 6 (4–8) | 6 (4–8) | 0.93 | 5.5 (4–8) | 6 (5–8) | 0.61 |
| 30-Day readmission | 33 (8) | 1 (2) | 32 (8) | 0.16 | 1 (2) | 5 (11) | 0.22 |
| 90-Day mortality | 18 (4) | 1 (2) | 17 (4) | 0.71 | 0 (0) | 2 (4) | 0.48 |
| Adjuvant therapy | 125 (29) | 17 (35) | 108 (28) | 0.31 | 15 (33) | 12 (26) | 0.70 |
Data are represented as n (%) or median (interquartile range). Percentages have been rounded and may not total 100.
Table 5:
Postoperative pulmonary complications
| Complication | All (n = 438) | Biologic (n = 49) | Synthetic (n = 389) |
|---|---|---|---|
| Pneumonia | 26 (6) | 5 (10) | 21 (5) |
| Empyema | 6 (1) | 2 (4) | 4 (1) |
| Haemothorax | 1 (0.2) | 0 (0) | 1 (0.3) |
| Acute respiratory distress syndrome | 2 (0.5) | 0 (0) | 2 (1) |
| Pneumonitis | 5 (1) | 0 (0) | 5 (1) |
| Pulmonary embolus | 4 (1) | 0 (0) | 4 (1) |
| Pleural effusion requiring reintervention | 12 (3) | 0 (0) | 12 (3) |
Data are represented as n (%). Percentages have been rounded and may not total 100.
After matching, surgical site complications requiring reoperation were observed in 4% (n = 2) vs 9% (n = 4) of patients (P = 0.68), prosthetic infection was observed in 0 vs 2% (n = 1) of patients (P = 1) and material fracture was observed in 0 vs 2% (n = 1) of patients (P = 1). One patient with prosthesis infection required the removal of the prosthesis, followed by vacuum-assisted closure and reconstruction with an anterior lateral thigh muscle. The 30-day complication rate was 39% (n = 18) vs 35% (n = 16) (P = 0.86), the pulmonary complication rate was 11% (n = 5) vs 15% (n = 7) (P = 0.77) and the grade ≥3 complication rate was 4% (n = 2) vs 13% (n = 6) (P = 0.29; Table 4).
On univariable logistic regression analysis of the matched cohort, extended chest wall resection with pulmonary resection [odds ratio (OR) = 0.34, 95% confidence interval (CI), 0.13–0.76; P = 0.008], history of radiation (OR = 2.51, 95% CI, 1.19–5.22; P = 0.017) and reoperation (OR = 2.76, 95% CI, 1.33–5.75; P = 0.007) were associated with surgical site complications requiring reoperation. Type of prosthesis (synthetic, OR = 1.24, 95% CI, 0.42–5.30; P = 0.73) and period of surgery (2011–2022, OR = 1.12, 95% CI, 0.54–2.38; P = 0.76) were not associated with surgical site complications requiring reoperation. On the multivariable analysis, operative time [adjustedOR (aOR) = 1.01, 95% CI, 1.00–1.01; P = 0.006] and operative blood loss (aOR = 1.00, 95% CI, 1.00–1.00; P = 0.012) were independently associated with a higher rate of surgical site complications requiring reoperation. Microvascular free flap reconstruction (aOR = 0.03, 95% CI, 0.00–0.42; P = 0.024) was associated with a lower rate of surgical site complications requiring reoperation.
The incidence of surgical site complications requiring reoperation between biologic and synthetic prosthesis was not significantly different when analyses were performed with titanium prostheses included in the biologic group (matched, n = 56; synthetic, univariable OR = 0.38, 95% CI, 0.05–1.84; P = 0.23; Q = 1.00, multivariable aOR = 0.26, 95% CI, 0.03–1.52; P = 0.14; Q = 0.70) or with the biologic group including only acellular dermal matrix reconstructions (matched, n = 19; synthetic, univariable OR = 3.37, 95% CI, 0.39–72.1; P = 0.28; Q = 1.00) (Supplementary Material, Tables S3 and S4).
DISCUSSION
The association between complete chest wall resection and increased survival rates has led to performing more extensive resections, which consequently require more complex reconstructions [1, 6, 9, 17, 21]. However, notwithstanding the wider range of prosthetic options available and the advances in reconstructive strategies, the concept of an ideal reconstruction remains elusive [11, 15, 21]. In this context, and in contrast to previous findings [15–17], our analyses demonstrated no significant differences between biologic and synthetic prostheses with respect to surgical site complications requiring reoperation, 30-day complications, pulmonary complications or grade ≥3 complications in a matched cohort. Additionally, microvascular free flap reconstructions were associated with a lower rate of surgical site complications requiring reoperation. These results may raise doubts about the pursuit of the ideal prosthesis and instead emphasize the need to continue to evolve towards patient-tailored strategies for full-thickness chest wall reconstruction that take into consideration the advantages and disadvantages of each type of prosthesis as well as each type of myocutaneous coverage.
The importance of restoring the anatomy and functionality of the chest wall (i.e. biomimesis) has been clearly demonstrated in trauma surgery [8], where stabilization of the flail chest is associated with fewer respiratory complications; consequently, the avoidance of flail chest is also an important aim after chest wall resection [5, 7]. Such findings led to the development of reconstruction algorithms in the setting of chest wall resection that emphasize the importance of reconstructing anterior defects, posterior defects not covered by the scapula and defects likely to produce paradoxical breathing movement [9]. At present, synthetic prostheses contoured to the chest wall continue to be the predominant materials used to cover chest wall defects [7, 22]. However, synthetic prostheses seem more susceptible to local infection, which can occur in up to 23% of patients after chest wall reconstruction and frequently leads to the removal of the implant [5, 7]. Biologic prostheses were specifically developed to address this issue, and both in vitro and in vivo data demonstrate a lower risk of bacterial adherence [9, 13, 15]. Additionally, biologic prostheses are remodelled into the surrounding tissue, which results in a more cohesive integration, while their lower immunogenicity results in less adherence to underlying organs, compared to synthetic prostheses [10, 23].
In agreement with previous studies [9, 15, 17], prosthesis infections were not observed in the biologic group in our study. Since the observed prosthesis infection rate was only 2% in the synthetic group, we were not able to detect a statistically significant difference between materials. Therefore, it can be reasoned that prosthesis infection does not depend only on the type of prosthesis used. Other factors, such as patient characteristics and comorbidities (age, diabetes, etc.), surgical management with attention to myocutaneous coverage and postoperative care, may play a substantial role in the development of prosthesis infection and local sepsis [9, 19, 24].
The statistically significant difference in surgical site complications by prosthesis type observed in the study from Giordano et al. [15] (16% after biologic reconstruction and 33% after synthetic reconstruction) could therefore be the result of selection bias that was not completely corrected by the multivariable analysis. By virtue of the matched correction for a large number of characteristics associated with postoperative outcomes [4, 7, 9, 25], and the consequent reduction in selection bias, our results demonstrated that the development of surgical site complications requiring reoperation was not associated with the type of prosthesis used. This finding is further supported by the observation that the risk (i.e. odds ratio) of surgical site complications requiring reoperation did not change significantly when the composition of the biologic group changed (adding titanium or including only acellular dermal matrix). Adequate healing may be influenced by various risk factors (age >75 years, chronic obstructive pulmonary disease, diabetes, steroid use), sufficient oxygen delivery and tension-free closure [26]. We believe that a bulky, well-perfused, tension-free myocutaneous coverage can prevent surgical site complications and, consequently, protect the underlying synthetic prosthesis from bacterial infiltration. This hypothesis is supported in our study by the lower rate of surgical site complications requiring reoperation when microvascular free flaps were used. In fact, our group also recently demonstrated that the use of microvascular free flaps allows larger defects to be covered without major postoperative morbidity [19].
Postoperative and pulmonary complications after chest wall resection are observed more frequently after extended resections with pulmonary resection [7, 25]. Despite the large proportion of extended resections with pulmonary resection performed in our study (44%), the overall rate of pulmonary complications (13%) was comparable to rates observed in previous studies [7, 15, 17, 25]. Our results, which demonstrate no difference in postoperative pulmonary complications between biologic and synthetic prostheses—combined with those from Weyant et al. [7] and Spicer et al. [25], which demonstrate no difference between rigid and flexible prostheses—suggest that the low rate of pulmonary complications described in current literature is determined less by the type of prosthesis but rather by the quality of full-thickness reconstruction aimed at restoring biomimesis. While associated with better breathing mechanics [12, 27], the influence of 3D reconstructions on the pulmonary complications rate will require further investigation. In fact, our experience with reconstructions shaped to contour the chest wall (such as the biosandwich, arena roof and neo-rib techniques) already demonstrates low rates of pulmonary complications [7, 21, 22, 28].
Finally, the use of biologic prostheses comes with a higher capital cost. Depending on the size of the prosthesis, the cost of synthetic prostheses ranges from $500 to $3000, whereas the cost of biologic prostheses ranges from $3000 to $30 000 [29, 30]. In addition, cost-effectiveness estimates should include the costs related to complications, reoperations and multistage strategies. Such cost-effectiveness analysis requires a well-designed prospective protocol and was beyond the scope of this study. Nevertheless, the higher capital cost of these materials should be taken into consideration when deciding on the type of prosthesis.
Despite our efforts to decrease bias and the influence of confounding variables by performing matched analysis, retrospective studies inherently remain subject to selection and reporting bias. In this setting, this study represents the experience on chest wall reconstruction from a single high-volume cancer centre with specific patient characteristics, sample size and outcomes. However, the limited sample size, combined with the small effect size, needs to be taken into consideration when interpreting the results. To the best of our knowledge, this is the largest and most comprehensive comparison of the use of biologic and synthetic prostheses in chest wall reconstructions in a matched cohort. Moreover, the use of matching allowed us to approximate randomization, which increases the clinical relevance and generalizability of our findings.
CONCLUSION
In our experience, surgical site complications requiring reoperation were not significantly different between biologic prostheses and synthetic prostheses in chest wall reconstructions. The observed lower rate of surgical site complications requiring reoperation after microvascular free flap coverage in our cohort could highlight the importance of an adequate full-thickness reconstruction of the chest wall defect. Rather than focusing on the search for an ideal prosthesis, our study confirms the need for a patient-tailored strategy that is based both on the choice of the prosthesis and on an adequate myocutaneous coverage of the extensive chest wall defect, aimed at reducing surgical site complications.
Supplementary Material
ACKNOWLEDGEMENTS
We thank the patients and their families, without whom this research would not be possible. We also thank the faculty members and nursing staff of Memorial Sloan Kettering Cancer Center who provided and cared for these patients. Finally, we thank David B. Sewell of the Memorial Sloan Kettering Department of Surgery for his superb editing assistance.
Glossary
ABBREVIATIONS
- CI
Confidence interval
- ECOG
Eastern Cooperative Oncology Group
- IQR
Interquartile range
- OR
Odds ratio
Contributor Information
Stijn Vanstraelen, Thoracic Service, Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
Manjit S Bains, Thoracic Service, Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
Joe Dycoco, Thoracic Service, Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
Prasad S Adusumilli, Thoracic Service, Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
Matthew J Bott, Thoracic Service, Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
Robert J Downey, Thoracic Service, Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
James Huang, Thoracic Service, Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
James M Isbell, Thoracic Service, Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
Daniela Molena, Thoracic Service, Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
Bernard J Park, Thoracic Service, Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
Valerie W Rusch, Thoracic Service, Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
Smita Sihag, Thoracic Service, Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
Robert J Allen, Jr, Plastic Surgery Service, Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
Peter G Cordeiro, Plastic Surgery Service, Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
Michelle R Coriddi, Plastic Surgery Service, Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
Joseph H Dayan, Plastic Surgery Service, Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
Joseph J Disa, Plastic Surgery Service, Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
Evan Matros, Plastic Surgery Service, Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
Colleen M McCarthy, Plastic Surgery Service, Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
Jonas A Nelson, Plastic Surgery Service, Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
Carrie Stern, Plastic Surgery Service, Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
Farooq Shahzad, Plastic Surgery Service, Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
Babak Mehrara, Plastic Surgery Service, Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
David R Jones, Thoracic Service, Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
Gaetano Rocco, Thoracic Service, Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
SUPPLEMENTARY MATERIAL
Supplementary material is available at EJCTS online.
Funding
This work was supported, in part, by the National Institutes of Health/National Cancer Institute [Cancer Center Support Grant P30 CA008748].
Conflict of interest: Matthew J. Bott is a consultant for AstraZeneca. James M. Isbell has stock ownership in LumaCyte and is a consultant/advisory board member for Roche Genentech. Daniela Molena serves on a steering committee for AstraZeneca and as a consultant for Johnson & Johnson, Bristol Myers Squibb, Merck and Genentech. Bernard J. Park serves as a consultant for Intuitive Surgical, AstraZeneca, CEEVRA and Medtronic. Valerie W. Rusch reports grant support (institutional) from Genelux and Genentech, travel support from Intuitive Surgical and travel support and payments from NIH/Coordinating Center for Clinical Trials. Babak Mehrara is a consultant for PureTech and Mediflix and receives royalty payments from PureTech. He is also the recipient of investigator-initiated awards from PureTech, Regeneron and Pfizer. David R. Jones serves as a consultant for AstraZeneca and on a Clinical Trial Steering Committee for Merck. Gaetano Rocco has a financial relationship with Scanlan, AstraZeneca and Medtronic. The other authors have no conflicts of interest to disclose.
DATA AVAILABILITY
The data underlying this article will be shared on reasonable request to the corresponding author.
Author contributions
Stijn Vanstraelen: Data curation; Formal analysis; Writing—original draft. Manjit S. Bains: Writing—review & editing. Joe Dycoco: Data curation. Prasad S. Adusumilli: Writing—review & editing. Matthew J. Bott: Writing—review & editing. Robert J. Downey: Writing—review & editing. James Huang: Writing—review & editing. James M. Isbell: Writing—review & editing. Daniela Molena: Writing—review & editing. Bernard J. Park: Writing—review & editing. Valerie W. Rusch: Writing—review & editing. Smita Sihag: Writing—review & editing. Robert J. Allen: Writing—review & editing. Peter G. Cordeiro: Writing—review & editing. Michelle R. Coriddi: Writing—review & editing. Joseph H. Dayan: Writing—review & editing. Joseph J. Disa: Writing—review & editing. Evan Matros: Writing—review & editing. Colleen M. McCarthy: Writing—review & editing. Jonas A. Nelson: Writing—review & editing. Carrie Stern: Writing—review & editing. Farooq Shahzad: Writing—review & editing. Babak Mehrara: Writing—review & editing. David R. Jones: Writing—review & editing. Gaetano Rocco: Conceptualization; Formal analysis; Supervision; Writing—original draft.
Reviewer information
European Journal of Cardio-Thoracic Surgery thanks Larry R Kaiser, Luca Voltolini and the other anonymous reviewer(s) for their contribution to the peer review process of this article.
REFERENCES
- 1. Shewale JB, Mitchell KG, Nelson DB, Conley AP, Rice DC, Antonoff MB, et al. Predictors of survival after resection of primary sarcomas of the chest wall—a large, single-institution series. J Surg Oncol 2018;118:518–24. [DOI] [PubMed] [Google Scholar]
- 2. Jones GD, Caso R, No JS, Tan KS, Dycoco J, Bains MS, et al. Prognostic factors following complete resection of non-superior sulcus lung cancer invading the chest wall. Eur J Cardiothorac Surg 2020;58:78–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Wakeam E, Acuna SA, Keshavjee S.. Chest wall resection for recurrent breast cancer in the modern era. Ann Surg 2018;267:646–55. [DOI] [PubMed] [Google Scholar]
- 4. Seder CW, Rocco G.. Chest wall reconstruction after extended resection. J Thorac Dis 2016;8:S863–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Rocco G. Chest wall resection and reconstruction according to the principles of biomimesis. Semin Thorac Cardiovasc Surg 2011;23:307–13. [DOI] [PubMed] [Google Scholar]
- 6. Wang L, Yan X, Zhao J, Chen C, Chen C, Chen J, et al. Expert consensus on resection of chest wall tumors and chest wall reconstruction. Transl Lung Cancer Res 2021;10:4057–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Weyant MJ, Bains MS, Venkatraman E, Downey RJ, Park BJ, Flores RM, et al Results of chest wall resection and reconstruction with and without rigid prosthesis. Ann Thorac Surg 2006;81:279–85. [DOI] [PubMed] [Google Scholar]
- 8. Tanaka H, Yukioka T, Yamaguti Y, Shimizu S, Goto H, Matsuda H, et al. Surgical stabilization of internal pneumatic stabilization? A prospective randomized study of management of severe flail chest patients. J Trauma 2002;52:727–32. [DOI] [PubMed] [Google Scholar]
- 9. Rocco G, Martucci N, La Rocca A, La Manna C, De Luca G, Fazioli F, et al. Postoperative local morbidity and the use of vacuum-assisted closure after complex chest wall reconstructions with new and conventional materials. Ann Thorac Surg 2014;98:291–6. [DOI] [PubMed] [Google Scholar]
- 10. Rocco G, Fazioli F, Cerra R, Salvi R.. Composite reconstruction with cryopreserved fascia lata, single mandibular titanium plate, and polyglactin mesh after redo surgery and radiation therapy for recurrent chest wall liposarcoma. J Thorac Cardiovasc Surg 2011;141:839–40. [DOI] [PubMed] [Google Scholar]
- 11. Rocco G. Overview on current and future materials for chest wall reconstruction. Thorac Surg Clin 2010;20:559–62. [DOI] [PubMed] [Google Scholar]
- 12. Moradiellos J, Amor S, Córdoba M, Rocco G, Vidal M, Varela A.. Functional chest wall reconstruction with a biomechanical three-dimensionally printed implant. Ann Thorac Surg 2017;103:e389–91. [DOI] [PubMed] [Google Scholar]
- 13. Wiegmann B, Korossis S, Burgwitz K, Hurschler C, Fischer S, Haverich A, et al. In vitro comparison of biological and synthetic materials for skeletal chest wall reconstruction. Ann Thorac Surg 2015;99:991–8. [DOI] [PubMed] [Google Scholar]
- 14. Huang LZY, Elbourne A, Shaw ZL, Cheeseman S, Goff A, Orrell-Trigg R, et al. Dual-action silver functionalized nanostructured titanium against drug resistant bacterial and fungal species. J Colloid Interface Sci 2022;628:1049–60. [DOI] [PubMed] [Google Scholar]
- 15. Giordano S, Garvey PB, Clemens MW, Baumann DP, Selber JC, Rice DC, et al. Synthetic mesh versus acellular dermal matrix for oncologic chest wall reconstruction: a comparative analysis. Ann Surg Oncol 2020;27:3009–17. [DOI] [PubMed] [Google Scholar]
- 16. Gonfiotti A, Viggiano D, Vokrri E, Lucchi M, Divisi D, Crisci R, et al. Chest wall reconstruction with implantable cross-linked porcine dermal collagen matrix: evaluation of clinical outcomes. JTCVS Tech 2022;13:250–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. D’Amico G, Manfredi R, Nita G, Poletti P, Milesi L, Livraghi L, et al. Reconstruction of the thoracic wall with biologic mesh after resection for chest wall tumors: a presentation of a case series and original technique. Surg Innov 2018;25:28–36. [DOI] [PubMed] [Google Scholar]
- 18. Von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbrouckef JP.. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Bull World Health Organ 2007;85:867–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Vanstraelen S, Ali B, Bains MS, Shahzad F, Allen RJ, Matros E, et al. The contribution of microvascular free flaps and pedicled flaps to successful chest wall surgery. J Thorac Cardiovasc Surg 2023;166:1262–72. 10.1016/j.jtcvs.2023.05.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Zubizarreta JR, Paredes RD, Rosenbaum PR.. Matching for balance, pairing for heterogeneity in an observational study of the effectiveness of for-profit and not-for-profit high schools in Chile. Ann Appl Stat 2014;8:204–31. [Google Scholar]
- 21. Vanstraelen S, Moonsamy P, Bains MS, Shahzad F, Allen RJ, Matros E, et al Biosandwich technique for extensive chest wall reconstruction in patients with complex defects. JTCVS Tech 2023;18:164–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Suzuki K, Park BJ, Adusumilli PS, Rizk NP, Huang J, Jones DR, et al. Chest wall reconstruction using a methyl methacrylate neo-rib and mesh. Ann Thorac Surg 2015;100:744–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Holton LH, Chung T, Silverman RP, Haerian H, Goldberg NH, Burrows WM, et al. Comparison of acellular dermal matrix and synthetic mesh for lateral chest wall reconstruction in a rabbit model. Plast Reconstr Surg 2007;119:1238–46. [DOI] [PubMed] [Google Scholar]
- 24. Khalil HH, Malahias MN, Balasubramanian B, Djearaman MG, Naidu B, Grainger MF, et al. Multidisciplinary oncoplastic approach reduces infection in chest wall resection and reconstruction for malignant chest wall tumors. Plast Reconstr Surg - Glob Open 2016;4:1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Spicer JD, Shewale JB, Antonoff MB, Correa AM, Hofstetter WB, Rice DC, et al. The influence of reconstructive technique on perioperative pulmonary and infectious outcomes following chest wall resection. Ann. Thorac. Surg 102;2016:1653–9. [DOI] [PubMed] [Google Scholar]
- 26. Spiliotis J, Tsiveriotis K, Datsis AD, Vaxevanidou A, Zacharis G, Giafis K, et al Wound dehiscence: is still a problem in the 21th century: a retrospective study. World J Emerg Surg 2009;4:2–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Pontiki AA, Natarajan S, Parker FNH, Mukhammadaminov A, Dibblin C, Housden R, et al. Chest wall reconstruction using 3-dimensional printing: functional and mechanical results. Ann Thorac Surg 2021;114:979–88. [DOI] [PubMed] [Google Scholar]
- 28. Rocco G, La Rocca A, La Manna C, Martucci N, De Luca G, Accardo R.. Arena roof technique for complex reconstruction after extensive chest wall resection. Ann Thorac Surg 2015;100:1479–81. [DOI] [PubMed] [Google Scholar]
- 29. Miller DL, Force SD, Pickens A, Fernandez FG, Luu T, Mansour KA.. Chest wall reconstruction using biomaterials. Ann Thorac Surg 2013;95:1050–6. [DOI] [PubMed] [Google Scholar]
- 30. Herrero A, Gonot Gaschard M, Bouyabrine H, Perrey J, Picot MC, Guillon F, et al. Comparative study of biological versus synthetic prostheses in the treatment of ventral hernias classified as grade II/III by the Ventral Hernia Working Group. J Visc Surg 2022;159:98–107. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data underlying this article will be shared on reasonable request to the corresponding author.