Minimally Invasive Repair of Pectus Excavatum Without Bar Stabilizers Using Endo Close

Abstract

Objectives:

Since the introduction of the Nuss technique for pectus excavatum (PE) repair, stabilization of the bar has been a matter of debate and a crucial point for the outcome, as bar dislocation remains one of the most frequent complications. Several techniques have been described, most of them including the use of a metal stabilizer, which, however, can increase morbidity and be difficult to remove. Our study compares bar stabilization techniques in two groups of patients, respectively, with and without the metal stabilizer.

Subjects and Methods:

A retrospective study on patients affected by PE and treated by the Nuss technique from January 2012 to June 2013 at our institution was performed in order to evaluate the efficacy of metal stabilizers. Group 1 included patients who did not have the metal stabilizer inserted; stabilization was achieved with multiple (at least four) bilateral pericostal Endo Close™ (Auto Suture, US Surgical; Tyco Healthcare Group, Norwalk, CT) sutures. Group 2 included patients who had a metal stabilizer placed because pericostal sutures could not be used bilaterally. We compared the two groups in terms of bar dislocation rate, surgical operative time, and other complications. Statistical analysis was performed with the Mann–Whitney U test and Fisher's exact test.

Results:

Fifty-seven patients were included in the study: 37 in Group 1 and 20 in Group 2. Two patients from Group 2 had a bar dislocation. Statistical analysis showed no difference between the two groups in dislocation rate or other complications.

Conclusions:

In our experience, the placement of a metal stabilizer did not reduce the rate of bar dislocation. Bar stabilization by the pericostal Endo Close suture technique appears to have no increase in morbidity or migration compared with the metal lateral stabilizer technique.

Introduction

The Nuss technique is the most widely used approach for the treatment of pectus excavatum (PE). Over the years bar dislocation remains one of the main complications, although less frequent than in the past; Kelly et al.¹ described a reduction in bar dislocation rate from 12% to 1% over a 21-year period with the Nuss procedure and hypothesized that this was due mostly to the increased experience of surgeons. Park et al.² identified three types of bar displacement: (1) bar flipping (rotation of the bar), (2) lateral sliding (horizontal slipping of the bar), and (3) hinge-point disruption (dorsal shift of the bar), with type 1 being the most frequent mechanism of dislocation (Table 1).

Table 1.

The Park Classification of Bar Displacement

Type	Mechanism
Type 1	Bar flipping
Type 2	Lateral sliding
Type 3	Hinge-point disruption

The same authors, in a multivariate analysis, reported that the most relevant predictive factors of bar dislocation were depression severity and surgeons' learning curve period.³ In contrast, Fallon et al.⁴ reported that the Haller index was not associated with a higher incidence of bar migration.

Over the years, many bar stabilization techniques aimed to preventing bar displacement have been described.^5,6 In particular, to avoid bar flipping, some authors suggested the “three-point fixation method,” which consists in bar stabilization using pericostal sutures along the bar.⁷ Bar fixation with pericostal wires was previously described by Zallen and Glick,⁸ but stabilization was only on the right side with insertion of a clamp to grasp and twist the wire around the bar. Another recent study reported that a shorter bar is subjected to a lesser force by the anterior chest wall, with consequent reduced incidence of bar dislocation.⁹

One of the most popular techniques to avoid bar dislocation involved the use of a lateral stabilizer. Some authors have reported that the use of metal stabilizers (unilateral or bilateral) was associated with a decreased rate of bar migrations.⁴ However, Watanabe et al.¹⁰ described many local complications and morbidity in the postoperative period due to the metal stabilizer, which made bar removal more difficult.

A previous study by our group showed that removal of a bar without a metal stabilizer is faster and easier and can be accomplished with a smaller incision than removal of a bar with a metal stabilizer.¹¹ Moreover, in our clinical experience of 300 cases (authors' unpublished data), we observed 2 cases with severe flipping of the retrosternal bar (type 1 of the Park classification) despite bilateral placement of a metal stabilizer. For these reasons we aimed to compare the efficacy of the metal stabilizer and pericostal sutures in bar stabilization. Our purpose was to evaluate whether pericostal fixation could be an alternative to metal stabilizers in PE repair.

The hypothesis was that when pericostal sutures were used, the metal stabilizer did not give any additional advantages. In the literature, there are no reports comparing the role of metal stabilizer and pericostal Endo Close™ (Auto Suture, US Surgical; Tyco Healthcare Group, Norwalk, CT) sutures in avoiding bar displacement.

Subjects and Methods

A retrospective study was performed on a total of 57 consecutive patients with PE treated by the Nuss minimally invasive repair technique from January 2012 to June 2013 at our institution. Collected data included patient's demographics, Haller Index, number of bars placed, length of surgery, length of hospital stay, pain score (evaluated with a 0–10 numerical pain intensity scale and reported by the patient at discharge), complications, and length of follow-up.

Minimum follow-up period was 12 months. All the operations were performed by the same surgeon, by the standard Nuss technique, by right thoracoscopy and placement of Biomet^® (Biomet Inc., Global Home Group, Jacksonville, FL) bars.

Institutional board approval and informed consent, including details on stabilization methods, were obtained.

Depending on patient age, as well as PE morphology and severity, one or two bars were placed based on the surgeon's intraoperative decision. In the case of patients with a stiff thorax (older than 14 years) with severe PE extending up to the inferior sternal third, two bars were usually preferred.

We stabilized the bar in all patients with pericostal nonadsorbable 2/0 sutures (Ti-Cron™; Covidien, Mansfield, MA; obtained from Medline Industries, Inc., Mundelein, IL) placed using Endo Close under direct visualization using a 30° telescope on the right hemithorax (Fig. 1). If stabilization was achieved on both sides with at least one pericostal suture per side and a minimum of four total sutures, the metal stabilizer was not used, and patients were included in Group 1 (Table 2).

FIG. 1.

Right stabilization using Endo Close.

Table 2.

Patients' Demographics and Surgical Outcome for Group 1

									Endo Close
Patient	Age (years)	Haller Index	Symmetry	Duration (minutes)	Number of bars	Pain score	Hospital stay (days)	Follow-up (months)	Right	Left
1	23	4.9	No	110	1	2	6	12	4	4
2	16	4.1	No	105	2	4	7	12	4	4
3	18	4	Yes	45	1	4	6	12	3	1
4	18	3.4	No	110	1	1	7	12	3	2
5	14	3.2	No	65	1	3	7	12	2	1
6	15	4.6	No	55	1	4	7	12	3	2
7	16	5.6	No	75	1	5	6	13	3	2
8	19	3.8	No	95	2	4	5	13	6	4
9	23	4	No	215	2	6	12	13	4	4
10	15	3.8	Yes	50	1	7	7	14	3	2
11	6	4	Yes	75	1	4	9	13	4	1
12	16	9	No	50	1	7	8	13	3	2
13	14	4	No	55	1	7	10	13	3	3
14	19	5	Yes	95	2	5	5	13	6	5
15	21	3.9	Yes	95	2	3	13	16	4	5
16	23	3.2	No	95	2	4	6	16	4	4
17	20	3.2	No	105	2	3	7	18	6	3
18	15	3.6	Yes	55	1	4	7	17	3	2
19	11	6.1	No	60	1	8	6	17	4	2
20	17	3.9	Yes	100	1	3	7	16	3	2
21	16	4,8	No	90	2	4	7	14	4	4
22	15	4.8	No	80	2	10	7	15	6	3
23	16	3.2	No	70	1	4	5	15	3	2
24	20	3.4	No	50	1	8	7	20	2	2
25	16	4.8	No	210	2	4	6	16	6	2
26	7	4	No	85	1	4	7	17	3	1
27	16	3.9	No	55	1	5	5	17	2	1
28	14	2.9	Yes	95	1	5	8	18	2	3
29	17	3.7	No	100	1	4	6	19	3	3
30	17	4	No	55	1	4	6	19	3	2
31	11	4	Yes	50	1	6	7	19	3	3
32	18	2.8	No	65	1	6	7	21	3	3
33	18	3.2	No	60	1	5	6	20	3	3
34	14	5,7	No	50	1	6	7	20	3	3
35	16	3.2	No	100	2	7	7	20	6	4
36	9	3.1	No	60	1	4	5	12	3	2
37	18	3.4	Yes	70	1	4	6	16	4	2

If bilateral stabilization with pericostal sutures was considered by the surgeon not feasible and unsafe because of the risks related with the insertion of the Endo Close on the left hemithorax under a poor thoracoscopic view, one metal stabilizer was positioned on the left side, and patients were included in Group 2 (Table 3).

Table 3.

Patients' Demographics and Surgical Outcome for Group 2

Patient	Age (years)	Haller Index	Symmetry	Duration (minutes)	Number of bars	Pain score	Hospital stay (days)	Follow-up (months)	Right Endo Close
1	11	6.6	No	60	1	5	7	12	3
2	13	4	Yes	75	1	5	5	12	2
3	23	4.6	No	85	1	4	5	12	2
4	22	3.2	No	60	1	5	7	12	2
5	8	3.4	Yes	65	1	6	6	13	2
6	13	2.6	No	60	1	5	7	13	3
7	17	3.6	Yes	95	2	4	5	14	6
8	14	4.1	No	90	1	4	7	15	3
9	14	3.36	No	65	1	2	6	15	3
10	10	3.8	No	55	1	4	8	15	5
11	14	4	Yes	160	2	5	7	15	4
12	16	16	No	130	2	4	7	16	6
13	12	4.7	No	60	1	6	6	16	3
14	8	8	No	105	1	5	12	16	3
15	20	3.7	No	90	1	4	7	17	3
16	5	4.7	No	60	1	6	10	17	2
17	17	5.8	No	45	1	6	7	19	3
18	19	3.2	No	55	1	6	7	20	4
19	22	7.23	No	70	1	5	6	21	3
20	10	3.36	Yes	50	1	7	7	21	3

All patients received the same postoperative analgesia (epidural analgesia, morphine, and nonsteroidal anti-inflammatory drugs).

During the postoperative period and once discharged from the hospital, patients were encouraged to sleep in a supine position. Sports activities were allowed after a period of 3–6 months from surgery, but potentially traumatic sports were not permitted. All patients were followed up by the same surgeon who performed surgeries at 1, 6, and 12 months after surgery and then once a year. Radiological evaluations (two projections for chest x-ray) were performed in the case of a history of symptoms (describing symptoms such as pain and dyspnea), traumatic injuries, or evidence of morphological anomalies on physical examination, in order to identify a possible bar dislocation.

Statistical univariate analysis with the Mann–Whitney U test as appropriate was used to compare the quantitative parameters of the two groups. Bar dislocation rates between the two groups were compared using Fisher's exact test as appropriate.

Results

Fifty-seven patients were included in the study, and, in total, 71 bars were placed. Forty-eight bars were stabilized without a metal stabilizer; Group 1 contained 37 patients, whereas Group 2 contained 20 patients (23 metal stabilizers were placed in these patients). In Group 1, PE was asymmetric in 10 cases and symmetric in 27 according to the current morphologic classification^1,2; median number of pericostal sutures on the right side was 3 (range, 2–6), whereas the median number of pericostal sutures on the left side was 2 (range, 1–5). In Group 2, 5 patients had asymmetric PE, and 15 had symmetric PE; median number of pericostal sutures on the right side was 3 (range, 2–5).

Others surgical results and complications are reported in Table 4.

Table 4.

Surgical Intra- and Postoperative Results

Results	Group 1	Group 2	P
Median duration of surgery (minutes)	70 (range, 55–95)	65 (range, 60–90)	.79^a
Median Haller Index	4 (range, 3.2–9)	4 (range,3.2–8)	.47^a
Number of patients	37	20	.34^b
Number with two bars	11	3	.336^b
Median length of hospital stay (days)	7 (range, 6–7)	7 (range, 6–7)	.87^a
Median pain score	4 (range, 1–10)	5 ( range, 2–7)	.40^a
Median follow-up (months)	16 (range, 12–21)	15 (range, 12–21)	.94^a
Dislocations	0	2	.12^b

By Mann–Whitney U test.

By Fisher's exact test.

Using a univariate analysis (Mann–Whitney test), the two groups did not statistically differ in age at surgery, Haller Index, duration of intervention, pain score, and length of hospital stay. No difference between number of positioned bars was shown between Groups 1 and 2 using Fisher's exact test (P = .34). Surgical results and complications are reported in Table 4.

Placement of pericostal sutures required on average 5 minutes per side. There were no intraoperative complications in either group, nor were there any early postoperative dislocations of the bar. One patient in Group 1 (Patient 21) had a wound infection that was treated successfully with antibiotics and wound care. Two patients in Group 2 (Patients 14 and 18) experienced bar flipping dislocation at 4 and 10 months, respectively, and required a second operation. Fisher's exact test showed no statistically significant difference between the two groups in bar dislocation rate (P = .12).

The use of pericostal sutures without a stabilizer was not associated with a higher rate of major complications, bar migrations, re-admissions, and re-operations when compared with the use of the metal stabilizer.

Discussion

Although various bar stabilization techniques have been described in recent years, the incidence of bar displacement remained a significant complication after PE repair. In the literature, there is no agreement on significant risk factors for bar dislocation. The metal stabilizer was introduced by Nuss in 1998 and is still popular in many centers as the most effective fixation tool. However, in our experience, we observed some cases of bar dislocation in patients who received bar stabilization by a metal stabilizer on the left side and pericostal sutures only on one hemithorax. This could be explained by the fact that the metal stabilizer, on one or both sides, can fix only the extremity of the bar, avoiding lateral displacement but making bar flipping still possible. To avoid bar rotation, which is the most frequent type of dislocation, pericostal sutures placed along the bar on both sides could theoretically allow a better prevention. To our knowledge, stabilization with pericostal sutures alone without the use of any lateral stabilizer was described monolaterally only by Zallen and Glick,⁸ and several related complications (27%) were reported. The possibility of fixing the bar with pericostal sutures placed through thoracoscopy between the ribs on both sides is particularly appealing. For these reasons, in the last 2 years we have been progressively abandoning the use of the metal stabilizer when a satisfactory pericostal fixation could be achieved. As we previously demonstrated, it is possible to fix the bar on the left side using a right thoracoscopic approach and inserting the telescope under the bar through the mediastinal tunnel, even if it is not always easy.⁶

For the purpose of this study, we considered the period from January 2012 to June 2013 in order to have a sufficient length of follow-up. During this period, thanks to our increasing experience in bilateral pericostal bar fixation with Endo Close, we have observed a progressive increase in number of cases not requiring the use of the metal stabilizer. The overall complication rate in our series was 5%, which is comparable with the 5%–27% rates reported in the literature.¹²

The most common complication was bar migration, which occurred in 2 cases of our series, both with metal stabilizers.

Fisher's exact test showed no statistically significant difference between the two groups; therefore we could not demonstrate that pericostal sutures are better than the metal stabilizer in fixing the bar, although our study was perhaps underpowered due to the small number of patients. However, we found a trend toward decreased migration rates in the group with pericostal sutures compared with the group with the metal stabilizer, but we need a larger series to confirm this result.

Our results showed that the metal stabilizer is not a better tool to prevent bar dislocation compared with bilateral fixation using Endo Close.

Because the metal stabilizer can increase the morbidity of the Nuss procedure and complicate the removal of the bar,¹¹ we progressively abandoned its routine use. The use of the metal stabilizer could be more effective than pericostal sutures in preventing lateral dislocation of the bar (type 2 of the Park classification). This is theoretically a good point supporting the use of the stabilizer, but this mechanism of dislocation is very rare; in our overall experience in more than 300 cases, we observed a mild lateral dislocation in 1 patient only.

The limitations of our study include the fact that randomization was not performed and that it is a retrospective study.

A longer follow-up and a prospective randomized study are mandatory to validate these results. However, our clinical data suggest that the metal stabilizer can be avoided in most cases when bilateral stabilization with pericostal sutures is obtained. In most patients, pericostal fixation is feasible, safe, and at least as effective as the metal stabilizer in preventing bar displacement.

Footnotes

Acknowledgments

We thank Anna Capurro for her help in revising the manuscript.

Disclosure Statement

No competing financial interests exist.

References

Kelly

, Goretsky

, Obermeyer

, Kuhn

, Redlinger

, Haney

, et al. Twenty-one years of experience with minimally invasive repair of pectus excavatum by the Nuss procedure in 1215 patients. Ann Surg, 2010; 252:1072–1081.

Park

, Chung

, Lee

, Kim

. Mechanism of bar displacement and corresponding bar fixation techniques in minimally invasive repair of pectus excavatum. J Pediatr Surg, 2008; 43:74–78.

Park

HJ1

, Lee

. Complications associated with the Nuss procedure: Analysis of risk factors and suggested measures for prevention of complications. J Pediatr Surg, 2004; 39:391–395.

Fallon

, Slater

, Nuchtern

, Cass

, Kim

, Lopez

, et al. Complications related to the Nuss procedure: Minimizing risk with operative technique. J Pediatr Surg, 2013; 48:1044–1048.

Torre

, Jasonni

, Asquasciati

, Costanzo

, Romanini

, Varela

. Absorbable stabilisation of the bar in minimally invasive repair of pectus excavatum. Eur J Pediatr Surg, 2008; 6:407–409.

Torre

, Jasonni

, Asquasciati

, Costanzo

, Romanini

, Varela

. Bilateral endoclose approach for the stabilization of the bar in pectus repair. Eur J Pediatr Surg, 2008; 18:407–409.

Hebra

, Gauderer

, Tagge

, Adamson

, Othersen

Jr.

A simple technique for preventing bar displacement with the Nuss repair of pectus excavatum. J Pediatr Surg, 2001; 36:1266–1268.

Zallen

, Glick

. Miniature access pectus excavatum repair: Lessons we have learned. J Pediatr Surg, 2004; 39:685–689.

Ghionzoli

, Ciuti

, Ricotti

, Tocchioni

, Lo Piccolo

, Menciassi

, et al. Is a shorter bar an effective solution to avoid bar dislocation in a nuss procedure?. Ann Thorac Surg, 2014; 97:1022–1027.

10.

Watanabe

, Watanabe

, Obama

, Ohsawa

, Mawatari

, Ichimiya

. The use of a lateral stabiliser increases the incidence of wound trouble following the Nuss procedure. Ann Thorac Surg, 2004; 77:296–300.

11.

Torre

, Rapuzzi

, Rossi

, Carlucci

, Costanzo

, Jasonni

. Removal of pectus bar fixed with absorbable vs metallic stabilizers. J Pediatr Surg, 2011; 46:1338–1341.

12.

Castellani

, Schalamon

, Saxena

, Hoellwarth

. Early complications of the Nuss procedure for pectus excavatum: A prospective study. Pediatr Surg Int, 2008; 24:659–656.