Video-Assisted Thoracoscopic Surgery for the Nuss Procedure: A 5-Year Single-Institution Experience

Nguyen The May, M.D. (  nguyenthemay@gmail.com ) Viet-Tiep Friendship Hospital, Hai Phong 180000, Vietnam Nguyen Huu Uoc, M.D., Ph.D. Department of Cardiovascular and Thoracic Surgery, Viet Duc University Hospital, Hanoi 100000, Vietnam https://orcid.org/0000-0002-6981-9837 Pham Huu Lu, M.D., Ph.D. Department of Cardiovascular and Thoracic Surgery, Viet Duc University Hospital, Hanoi 100000, Vietnam Phung Duy Hong Son, M.D., Ph.D. Department of Cardiovascular and Thoracic Surgery, Viet Duc University Hospital, Hanoi 100000, Vietnam Vu Ngoc Tu, M.D., Ph.D. Department of Cardiovascular and Thoracic Surgery, Viet Duc University Hospital, Hanoi 100000, Vietnam https://orcid.org/0000-0003-4262-4471 Nguyen Viet Anh, M.D. Department of Cardiovascular and Thoracic Surgery, Viet Duc University Hospital, Hanoi 100000, Vietnam Khanh Nam Do, M.Sc. Institute for Preventive Medicine and Public Health, Hanoi Medical University, Hanoi 100000, Vietnam Kim-Duy Vu Institute for Preventive Medicine and Public Health, Hanoi Medical University, Hanoi 100000, Vietnam Hoang-Long Vo, M.D. (  vohoanglonghmu@gmail.com ) Institute for Preventive Medicine and Public Health, Hanoi Medical University, Hanoi 100000, Vietnam https://orcid.org/0000-0001-5992-875X Doan Quoc Hung, M.D., Ph.D. (  hung.doanquoc@gmail.com ) Hanoi Medical University, Hanoi 100000, Vietnam


Introduction
Since Nuss et al.'s report with 42 patients using a minimally invasive technique to correct the depression of the pectus excavatum (PE) in 1998 [1], this procedure (Nuss procedure) was considered the gold standard treatment for PE in prepubertal patients. The initial application of the Nuss technique was relatively limited; however, thanks to accumulated experiences across the world, the Nuss technique with several modi cations has been used extensively by many surgeons, such as adults [2], severe and/or asymmetrical deformities [3], and recurrent cases [4], [5], [6], [7]. In particular, the relative risk of complications signi cantly decreased, such as bar rotation, pneumothorax, and severe cardiopulmonary injury [5]. Various attempts have recently been made to reduce the risk of severe complications, including utilizing the originally designed sternum elevator or bar insertion with the extrathoracic approach [8], [9].
With universal acceptance, the additional use of the video-assisted thoracoscopic surgery (VATS) technique in the Nuss procedure demonstrated an increased safety of the operation with a decreased frequency of serious intraoperative and postoperative complications. Bufo et al. rst reported VATS for the prevention of life-threatening intraoperative lesions of the mediastinum [10]. The discussion is currently focused on the superiority of left, right or bilateral thoracoscopic access and the necessity of CO2 insu ation or sternal elevation to improve safety during a surgical procedure [11], [12], [13], [14].
Although serious life-threatening complications after the Nuss procedure are rare [15], [16], [17], VATS used during the operation may further reduce the perioperative risk. The available evidence in the literature was mostly reported from large institutions in developed countries, and no publications on the surgical outcomes and safety in the application of VATS for the Nuss procedure were known in resourcescare conditions such as Vietnam.
Therefore, this study was conducted to review our large single-center experience in Vietnam in the use of the VATS technique in the Nuss procedure for 365 consecutive patients with PE within the ve years and to evaluate mid-term to long-term outcomes in these patients.

Study Population
In this retrospective study, we consecutively selected patients with a diagnosis of PE who underwent VATS for the NUSS procedure from January 2015 to September 2019. A total of 365 patients who were surgically treated for PE at the Department of Cardiovascular and Thoracic Surgery, Viet Duc University Hospital (Hanoi, Vietnam) were included in the nal analysis.
The evaluation by complete history, physical examination, chest radiographs, electrocardiogram, pulmonary function test, echocardiogram, and computerized tomography (CT) of the chest were performed. The demographic data, surgical data, thoracoscopic ndings, early and late complications, length of hospital stay, and early, mid-term, and long-term outcomes were also collected. The indications for surgical repair were two or more of the following criteria demonstrated by Dr Nuss: (1) progression of the deformity; (2) exercise intolerance; (3) progressive chest pain or dyspnea; (4) restrictive ventilatory impairment; (5) Haller index > 3.25; (6) previous failed Ravitch procedure; (7) cardiac compression; and (8) mitral valve prolapse. All procedures were performed with the Nuss procedure, which included left, right or bilateral VATS.

Surgical techniques
All patients had a thoracic epidural for intraoperative anesthesia and postoperative pain control. The patient was placed in the supine position after general anesthesia. The arms of the patient were kept abducted approximately 70-80° in relation to the body (Figure 1).
In the determination of anatomic landmarks in surgery, we determined the concave area, the center of the concave area, the edges of the concave area, and the highest point of the concave edge on either side of the concave circumference. The metal bar was placed along a straight line that was formed from the center of the concave area and the two highest points of the concave edge ( Figure 2). The measurements of the thorax and the bending of the metal bar (sternum lift) were then performed ( Figure 3).
A 5-mm trocar was placed in the marked position on the left chest wall. After pumping low-pressure CO2 (5 mmHg) into the pleural cavity to collapse the lungs, the camera with a 30-degree optic was taken to the left chest to observe the entire chest and mediastinum during surgery. The skin incision was made above or below the position where the metal rod was intended to be placed in an intercostal space. One 2-cm vertical skin incision was made in the midaxillary line on each side (Figure 4). In the required cases with 2-bar insertion, only a skin incision was made between the two intended positions to place the two metal bars. Subcutaneous or submuscular dissections were performed from the cutaneous incision to the highest edge of the PE ( Figure 5).
Then, we created a tunnel through the mediastinum ( Figure 6). The heart-shaped clamp was used to pierce the left pleural space at the highest edge of the pit, slowly went close to the anterior chest wall towards the mediastinum at the deepest point of the pit, and gradually separated the pericardium from the posterior sternum. The introducer reaches the posterior surface of the sternum in the left pleural cavity and is further moved through the mediastinal pleura in the retrosternal space, next goes across the right pleural space and nally exits the chest through the right intercostal space. The electrocardiogram is continuously monitored during the process of creating a tunnel through the mediastinum.
During the reeving sternum lift bar stage, a Perlon thread was tied to the metal rod and then tied to the introducer. After the withdrawn introducer was withdrawn, the metal bar was threaded through the anterior mediastinum in a direction going from right to left under the control of thoracoscopy ( Figure 7). Then, the pectus bar was rotated to suit the patient's chest ( Figure 8).
The steel thread was sewn around the ribs and was then tied to the top of the metal bar ( Figure 9). Under the observation of endoscopy, the process of sewing steel threads was very safe, avoiding stitching into the lungs and checking bleeding. We did not use screw braces to x the metal bar as in the original Nuss surgery. Finally, the anesthesiologist squeezed the balloon through the endotracheal tube to allow the expanded lungs while gradually withdrawing the trocar and removing air from the bilateral pleural space.
A pleural drainage catheter was not placed.

Statistical Analysis
We conducted all analyses with Stata® 15 (StataCorp LLC, USA) for Windows. Continuous variables were expressed as the mean and standard deviation (SD) with interquartile ranges (IQR). Categorical variables are presented as counts with percentages. No imputation was made for missing data. A p-value<0.05 was considered to be statistically signi cant.

Ethical Approval
All procedures conducted in studies involving human subjects were in compliance with the ethical principles of the Institutional and/or National Study Committee and the Helsinki Declaration of 1964 and its corresponding amendments or equivalent ethical standards. This study was approved by the Ethics Board of Hanoi Medical University. All individual participants included in the study obtained informed consent.

Results
Preoperative clinical characteristics of PE patients who underwent VATS-NUSS Table 1 shows the preoperative clinical characteristics of 365 PE patients undergoing VATS for the NUSS procedure. The median age at operation was 15.61 ± 3.73 years (range, 5 to 27 years). A total of 84.65% were male. PE was commonly detected at puberty (n = 328, 89.9%). Underweight (BMI <18.5) was found in 84.4% of the patients, and no patient was recorded as overweight or obese. A total of 1.37% of the patients experienced previous PE repair surgery (n = 5). Of the 365 patients, 288 had symmetric PE (78.90%), and 77 had asymmetric PE (21.10%). Additionally, in Table 1, the numeric distribution of each morphologic type and subtype of pectus is described in detail.
Bar removal after VATS-NUSS Table 4 also indicates the outcomes related to the bar removal that was carried out in 175 patients (47.95%). The time the bar was in situ ranged from 2 to 49 months (mean: 28.89±7.48 years). In 3 patients (1.71%), bar removal was performed early, less than 1 year, due to metal bar allergy. We recorded a mean operative time at bar removal of 34.09 ± 10.61 minutes and a length of hospitalization following bar removal of 2.4 ± 1.34 days. Of 175 patients who underwent bar removal, the most frequent complication was pneumothorax, which was diagnosed in 19 (10.85%) patients; surgical wound infection in 1 patient and incision uid accumulation in 1 patient were observed following bar removal.

Discussion
In resource-scare conditions such as Vietnam, it di cult us to acquire all important data for a large cohort of consecutive PE patients who underwent VATS for the NUSS procedure. Hence, we believe that current rare evidence drawn from this cohort enables us to give a real picture in clinical practice valuably not only for Vietnam but also for countries with similar resource-scare conditions. As is known in the literature, proper metal bar shape is crucial in the Nuss procedure for patients with PE. Kelly R. E et al. (2010) favored a gentle semicircular curve laterally with a 2 to 4 cm at segment in the center to support the sternum [19]. Slight overcorrection of depression is preferred to undercorrection because the authors believed it minimizes the risk of recurrence and decreases the risk of cartilage buckling [19]. Park et al favor an asymmetrically bent bar for patients with asymmetric depression [20]. In this cohort, we used a ruler bent to the desired thoracic shape and then bent the metal bar to the shape and size of the ruler. We begin to bend the middle of the bar gradually to its ends; bars are usually taken to be 1 inch smaller compared to the distance between the two middle armpit lines. We have found that symmetrically bent bars tend to be unstable for the patients' chest.
In our series, most patients received only 1 bar (n = 350, 95.89%), while 15 patients received 2 bars (4.11%). The 2-bar insertion was indicated for most elderly patients with severe PE and wide concave area; no case was placed with 3 metal bars. Our nding was consistent with previous reports when both revealed that elderly patients need to put more metal bars compared to younger patients [21], [22]. The number of patients receiving 2 metal bars was lower in our study than in previous studies [21], [22]. For instance, in an early report by Pilegaard HK, 121 patients were inserted with 1 metal bar (66.8%), 57 cases were inserted with 2 metal bars (32%), and 2 cases were inserted with 3 metal bars (1.2%) [22]. Of 1006 PE patients in Krystian Pawlak's recent series from July 2002 to September 2016, one bar was implanted in 452 (44.9%) patients, two bars were inserted into 545 (54.2%) patients, whilst three bars were inserted into 9 (0.9%) patients [21]. The patient's chest wall became more solid after puberty and harder to bend; hence, the insertion of 2 metal bars was commonly shown with better and more stable results. Previous authors suggest that PE patients with Marfan syndrome or grand canyon PE and asymmetrical PE require two metal bars. The rate of using 2 metal bars has increased gradually in recent years; in particular, the insertion of 3 metal bars for PE patients has been reported in several studies. In the treatment of PE, the stability of the metal bar in the thorax is improved, and there is no need for re-surgical intervention when 2 metal rods are implanted in the rst surgery. The use of two metal bars during the rst treatment has become the standard of minimally invasive surgery for PE [1], [23], [24].
Previous reports have shown the safety of the thoracoscopic Nuss procedure with a decreased frequency of serious intraoperative and postoperative complications, especially in patients with complex PE who experienced a previous history of thoracic surgery and recurrent PE [18], [10], [12]. Usually, the surgeon stands to the right of the patient using the supportive thoracoscopic approach, which is consistent with the classic Nuss procedure that chooses the entrance from the right chest wall. The VATS technique was acquired for all PE patients in this cohort, and CO2 insu ation was used to create an arti cial pneumothorax and thereby increase the empty pleural space. Forty-three patients underwent right VATS (11.78%), while left VATS was performed in 308 patients (84.38%). In particular, a bilateral thoracoscopic approach was applied in 14 PE patients (3.84%) due to severe PE compressing the lungs and heart and failed Ravitch procedure. In the early years of implementing VATS in the Nuss procedure, we applied both thoracoscopic approaches, including the right chest wall and the left chest wall. Nevertheless, during recent years, most of the cases were approached through left thoracoscopy in our institution. With the left thoracoscopic approach, we found that the surgical eld of view was wide enough to clearly observe the entire pleural cavity, left lung, blood vessel, heart, pericardium, mediastinum and diaphragm; hence, serious complications can be avoided. Several authors favored placing the trocar in the middle axillary line [14], [25]. However, we found that the trocar placed in the middle axillary line can result in vision that can be hindered by the heart, lungs, and diaphragm, especially in cases of asymmetrical severe PE. We agree with Hendrickson et al. and Palmer et al. that left thoracoscopy has an advantage over right thoracoscopy and does not increase the risk of heart injury [11], [13]. Hemostasis is well controlled, and lung damage is avoided under endoscopic observation. During the process of ballooning through the endotracheal tube to enlarge the bilateral lungs, we clearly observed that the two lungs were enlarged and until the lungs were maximized; when the pleural space was empty, we withdrew the endoscopic trocar. Therefore, a pleural drainage catheter was not placed during surgery.
During the surgery, no serious complications or fatalities were observed in our series. Early postoperative complications were recorded up to 30 days during the hospital stay, with the most frequent pneumothorax occurring in 5 patients (1.37%). In most of the studies, the most common early complication was pneumothorax, and its occurrence frequency was estimated to be from 1% to 64% [12], [26], [27], [28]; however, in the majority of cases, the pneumothorax spontaneously resolved, and only a small number of patients required surgical treatment. In addition, no patient with postoperative bleeding required a blood transfusion, and no other serious complications or deaths were recorded in our study.
Late postoperative complications are events that occur when the patient is discharged from the hospital, such as metal bar deviation, metal bar allergy, or surgical wound infection. Regarding late postoperative complications, we recorded surgical wound infection in 2 patients (0.55%), metal bar deviation in 5 patients (1.37%), metal bar allergy in 10 patients (2.74%), and recurrent PE in 2 (0.55%). Importantly, no mortality was reported in this series. Bar displacement is one of the most common complications [29].
Metal bar allergy occurred in 2.74% of our patients (n = 10) with manifestations such as mild fever, in ammation, wound uid buildup, blisters, and erythema.  [19]. In particular, we found that 6 out of 10 patients with metal bar allergies had a history of allergic diseases such as allergic rhinitis, drug allergies, weather allergies, food allergies, and asthma. This helps us gain more experience in the prognosis, diagnosis, and treatment of allergy drugs as well as closely monitor allergy status after surgery for appropriate management measures.
The Nuss procedure is known as a prepared clean surgery. Although very few reported wound infections have occurred, infections are extremely serious because foreign material has been placed inside the body.
In our study, of 365 postoperative patients, there were 2 cases of late wound infection (0.55%). Those patients received prompt antibiotic treatment and were discharged after 7-10 days of treatment. There were no cases of deep infection of the metal bar. Some studies in the literature showed that the incidence of late wound infection was 1.5% [32], [33]. In addition, other complications should also be noted in our report. Five patients had chest wall pain lasting at 1 month and 2 months postoperatively, were taken oral pain relievers and performed breathing exercises; their pain then subsided and disappeared. There were 2 patients with recurrent PE and 5 with persistent PE after the bar removal operation.
We Several limitations need to be noted in this study. It was a retrospective one-center experience. Although 60.27% of patients were followed up over 30 months, mid-term to long-term postoperative outcomes after removal of bar(s) should be investigated in detail in a further study. In particular, metal support is still in place in a number of PE patients, not enabling us to have the complete evaluation of postoperative results.

Conclusions
In conclusion, the VATS technique in the Nuss procedure for PE was a safe and effective approach that minimized the occurrence of serious intra-and postoperative complications. From our initial oneinstitution experience in a resource-scare country, good mid-term to long-term postoperative outcomes in PE patients were obtained with VATS treatment using the minimally invasive Nuss method. Current rare evidence drawn from this cohort enables us to provide a real picture of the application, modi cation and development of VATS in the Nuss procedure not only for Vietnam but also for countries with similar resource-scarce conditions.

Declarations
Ethical Approval All procedures conducted in studies involving human subjects were in compliance with the ethical principles of the Institutional and/or National Study Committee and the Helsinki Declaration of 1964 and its corresponding amendments or equivalent ethical standards. This study was approved by the Ethics Board of Hanoi Medical University. All individual participants included in the study obtained informed consent.
Competing Interests: The authors declare no competing interests. The determination of anatomic landmarks in surgery  The creation of skin tunnel in the left chest wall Figure 6 The process of created tunnel through the mediastinum under the control of the thoracoscopy Page 24/24 Figure 7 The process of created tunnel through the mediastinum under the control of the thoracoscopy Figure 8 Bending and rotating the metal bar Figure 9 The pectus bar was anchored into the position by steel thread