Determining the Optimal Technique for Bar Fixation in the Repair of Pectus Excavatum

Introduction

Pectus excavatum, the most common chest wall deformity, is a condition in which the rib cage pushes the sternum toward the spine to create a central depression. ¹ Surgical correction is commonly performed through a minimally invasive technique by inserting a metal bar beneath the sternum, immediately correcting the deformity.^2,3 One of the major variables in the variations of the technique is how the bar is secured. A common means of fixation is the use of metallic bar stabilizers.^3,4 However, their use has been associated with wound complications and increased difficulty in removing the bar after treatment completion.^3,5

Furthermore, pain associated with the stabilizer site requiring early stabilizer removal has been previously reported.^6,7 Metal stabilizers are used mainly to prevent bar displacement, nonetheless, recent studies fail to demonstrate a reduction in the rate of bar dislocation, even with bilateral stabilizers. ⁴ As minimally invasive repair of pectus excavatum has become a common procedure performed by pediatric surgeons, some have been moving away from using bilateral stabilizers to using unilateral or none at the time of bar placement.

Because of the reported complications with the use of metal stabilizers, we aimed to determine the optimum technique for bar fixation by reviewing our institutional experience with the use of bilateral, unilateral, and no stabilizer placement. We hypothesized that patients with bilateral stabilizers will have higher rates of stabilizer complications with no difference in the rate of bar displacement when compared with those with unilateral or no stabilizers.

Materials and Methods

After institutional review board approval, all patients who underwent minimally invasive bar placement and subsequent bar removal were identified between December 2001 and July 2019. Patients were excluded if they had an open repair, more than one bar, no documented follow-up, or incomplete data regarding the number of stabilizers or bars placed during surgery. A retrospective review of chart data was performed to identify demographic data, the Haller index, operative details, and postoperative follow-up details. Complications included pain associated with the stabilizer site, surgical site infections (SSIs), and bar displacement. SSI was defined as evidence of erythema, swelling, abscess formation, or purulent discharge at the surgical incision site requiring drainage and/or antibiotic therapy.

The surgical technique in all cases included a subxiphoid incision to guide the bar across the sternum through the surgeon's finger as previously described. ⁸ The pediatric surgeons at our institution do not use thoracoscopy for bar placement nor the sternal lift. All bars are stainless steel unless metal allergy screening and testing were positive before surgery. In those cases, the patient gets a titanium bar. Cryoablation is used as part of our standard surgical practice in the repair of pectus excavatum. Our cohort had 3 groups of patients: those with bilateral stabilizers, those with unilateral stabilizers, and those without stabilizers. In patients without a stabilizer, the bar ends were secured to the muscle using an absorbable suture, such as Vicryl, and stabilizers were fixed using a polyester nonabsorbable suture.

We have no predefined criteria to determine stabilizer use. This decision was made by the attending surgeon based on his or her preferences and how the bar fits between two ribs once in place. Postoperative follow-up occurred between weeks 2 and 4, then at 3 months, and as needed until ready for bar removal at ∼3 years.

Study data were collected and managed using REDCap electronic data capture tools.^9,10 Data are presented as medians with interquartile ranges (IQRs) and frequencies with percentages. STATA 15 (StataCorp 2017. Stata Statistical Software: Release 15; StataCorp LLC, College Station, TX, USA) software was used for statistical analyses.

Results

A total of 582 patients met the inclusion criteria of whom 21 were excluded because they had two bars placed during surgery. The cohort was predominantly male with a median age at the time of bar placement of 15 years (IQR 12.4, 16.3) (Table 1). The majority had bilateral stabilizers placed (91.3%, n = 512), followed by unilateral stabilizers (8.4%, n = 47) and no stabilizers (0.4%, n = 2). The median time to bar removal was 3.0 years (IQR 2.9, 3.1).

Pain attributed to the stabilizer site that required removal was observed only in the bilateral stabilizer group (Table 2). SSI related to the stabilizer site occurred with both bilateral and unilateral stabilizers groups, whereas bar displacement was observed only with bilateral stabilizers (Table 2). There were no documented complications in the no stabilizer group.

Discussion

In this single-center study, we observed that as the trend moves toward unilateral and no stabilizer use in the repair of pectus excavatum, there were fewer cases of pain requiring surgical intervention with no increased number of bar displacements. However, SSIs were observed with similar frequency in patients with bilateral and unilateral stabilizers.

Metal stabilizers have been reported to cause pressure points in the skin that can result in erosion, seroma, infection, and pain.^5–7 Although not statistically significant, we observed a higher frequency of complications associated with bilateral stabilizers (4.9%) than unilateral stabilizers (2.1%). It is reasonable that the likelihood of having complications related to the stabilizers increases with the number of stabilizers placed. Avoiding stabilizers results in less local discomfort, reduced procedure costs, and shorter operative time, and facilitates subsequent bar removal. ¹¹

All cases of bar displacement in our cohort were in patients who had bilateral metal stabilizers. However, with only 47 patients in the unilateral stabilizer group, there can be little confidence in this raw difference to date. According to the rule of zero numerators, there should be a maximum displacement rate of 6% based on the current sample. ¹² We will need much more experience to see whether this holds up. However, it should be noted that our displacements have all occurred earlier in our experience with none occurring in the past 8 years. This is the same timeframe that we have been using fewer stabilizers. Our hypothesis would be that we are bending the bar to a tighter fit such that the stabilizers have less of a role.

Bar displacement in patients with bilateral stabilizers has been reported before, demonstrating their presence does not prevent bar movement, and other factors need to be considered. ⁴ Patients with high Haller index and those with reduced chest wall malleability have been reported to be susceptible to bar displacement due to the elevated pressures the bar is exposed to in these patients. ¹³ In addition, using shorter bars can reduce tension on the sutures applied at bar extremities, leading to enhanced bar stability and a reduced risk of bar displacement. ¹⁴

The three bars were displaced ∼45° requiring reoperation to reposition the bar. Two of the patients with bar displacement in our cohort also had an SSI. One had the SSI after the bar was repositioned and the other patient had it concurrently with the bar displacement, which was repositioned after the SSI resolved. Both were treated with abscess drainage and antibiotics.

This single-center retrospective study has several limitations. First, the study may be affected by a lack of generalizability, as the findings may not apply to other populations or settings. Its retrospective nature makes it prone to confounding variables that cannot be controlled for. In addition, the small number of patients in the unilateral and no stabilizer groups compared with the bilateral group makes statistical comparisons unreliable as it is more difficult to detect meaningful differences between the groups, allowing only a descriptive analysis of the results.

The small number of patients with no stabilizers compared with those with stabilizers makes it difficult to conclude that avoiding stabilizers is associated with a reduction in complications; however, existing evidence and our experience continue to suggest limited utility and more complications associated with the use of metal stabilizers. Despite these limitations, this descriptive study shows that bilateral stabilizers are associated with a range of problems and suggests that moving toward the use of unilateral or no stabilizers might lessen wound complications without an increased risk of bar displacement. We will continue to follow the current cohort who have the bar in place since most of these patients have unilateral stabilizers.

Conclusion

As the trend moves toward unilateral and no stabilizer use, we observe fewer cases of pain requiring stabilizer removal. In our early institution experience, unilateral stabilizer use for bar fixation in the repair of pectus excavatum appears safe, with no increased risk of bar displacement.