A bioresorbable osteosynthesis device can induce an earlier sternal fusion after median sternotomy

Abstract

OBJECTIVES

We examined the impact of the bioresorbable osteosynthesis sternal pin (Super Fixsorb 30) on sternal healing after median sternotomy.

METHODS

Sixty-three patients who underwent aortic surgery through median sternotomy between January 2006 and March 2009 were analysed. Sternal pins were utilized in 36 patients in addition to the standard closure of the sternum with Ethibond sutures (Group A), and 27 patients received no pins with the standard Ethibond sternal closure (Group B). The occurrence of transverse sternal dehiscence, anterior–posterior displacement and complete fusion of the sternum were evaluated by a computed tomography scan. The cross-sectional cortical bone density area (CBDA) of the sternum was examined to evaluate the osteoconductivity of the sternal pin over a 12-month period.

RESULTS

There was no sternal displacement (0%) observed in Group A at discharge. Meanwhile, five displacements (18.5%) were observed in Group B (P = 0.007). The complete sternal fusion rates at 12 months postoperatively were 100% in Group A, and 21.6% in Group B (P < 0.001). A significant increase in the CBDA was observed in Group A (P < 0.001; between CBDA at discharge and 12 months postoperatively).

CONCLUSIONS

The Super Fixsorb 30 sternal pin reduced an anterior–posterior sternal displacement and facilitated an earlier sternal fusion. The pin may have the potential to promote osteogenesis.

INTRODUCTION

Sternal instability can result in sternal infection after a median sternotomy [1]. Although numerous methods for sternal osteosynthesis have been described [2, 3], the current standard for sternal closure remains circlage wire fixation. Several reports have been presented regarding the efficacy of poly-l-lactide (PLLA)-containing pins in facilitating sternal fixation [4, 5]. However, the PLLA-only material has disadvantages such as the lack of osteoconductivity and the lack of the ability to fuse with the bone. A new bioresorbable osteosynthesis device (Super Fixsorb 30; Takiron Co., Ltd, Osaka, Japan), which is a mixture of PLLA and particulate resorbable uncalcined hydroxyapatite (u-HA) (u-HA–PLLA), is considered to have a higher mechanical strength, total resorbability and osteoconductivity [6–8]. Since 2006, we have utilized the sternal pin for sternal closure after a median sternotomy. In this study, we retrospectively investigated the effects of the sternal pin on sternal healing by examining the serial computed tomography (CT) scans of patients with or without the sternal pin for sternal closure following aortic surgery.

MATERIALS AND METHODS

This study was approved by the Ethics Committee, which waived the need to obtain patient consent.

The subjects were 63 patients (age; mean: 67.1 ± 9.85 years, range 38–83 years, 47 males) who underwent aortic surgery through median sternotomy between January 2006 and March 2009. The subjects included 49 patients with atherosclerotic aneurysms and 14 with aortic dissections. Fifty-eight patients had ascending aorta and total aortic arch replacements, and five patients had ascending aorta and hemi-aortic arch replacements.

Surgical techniques for sternum closure

A full median sternotomy was closed with number-5 Ethibond non-absorbable braided polyester sutures (Ethicon, Inc., Somerville, NJ, USA) with or without Super Fixsorb 30 sternal pins. We prefer these sutures to the standard stainless steel wires for sternal closure for fear of tearing the sternum with the wires. The patients were divided into two groups according to the application of the pin for sternal closure (Group A: with the pin; Group B: without the pin). The sternal pin was randomly applied, mainly based on surgeons' preferences. In Group A, after number-5 Ethibond sutures were applied to the sternum in a figure of eight fashion, two small holes were made on the edge of the two pieces of the sternum with a dedicated reamer; one at the manubrium and the other at the sternal body. The pins were inserted into the holes with a dedicated applier (Supplementary Video 1). This procedure usually required <1 min per pin.

Supplementary Video 1: Sternal closure with Super Fixsorb 30 sternal pins.

Computed tomographic evaluations

After aortic surgery, the follow-up CT scans were routinely conducted at discharge and 3, 6, 12 and 24 months postoperatively, primarily to evaluate late aortic complications. In the present study, we utilized these serial follow-up CT scans for evaluations of the occurrence of sternal dehiscence, displacement and complete sternal fusion. Sternal dehiscence was defined as a disruption of the two pieces of the sternum in a lateral direction throughout the entire length of the sternum (Fig. 1a). Sternal displacement was defined as a misalignment of the two pieces of the sternum >50% of the sternal width in an anterior–posterior direction (Fig. 1b), and complete sternal fusion was defined as the complete fusion of the sternum throughout the entire length of the sternum.

Figure 1:

CT images of (a) sternal dehiscence and (b) sternal displacement.

Cross-sectional cortical bone density area of the sternum

To evaluate the effect of the sternal pin on sternal osteoconductivity, the CBDA was measured on a sample slice of the CT scan, and time-dependent changes were examined. In Group A, the sample slice chosen was the slice where the sternal pin could be visualized in its entirety. The same sample slice level was utilized in every evaluation. In Group B, the sample slice was chosen as one random slice, and the same sample slice level was utilized in every evaluation. As the sternal cortical bone was clearly distinguished from soft tissues from the outside and cancellous bone was clearly distinguished from the inside by the CT scan, tracing the boundary of the sternal cortical bone was done manually by means of Rapideye DICOM Viewer (Toshiba medical systems, Tokyo, Japan). All CT scans were conducted at a 5 mm width interval.

Statistical analysis

Data analyses were performed with SPSS 17.0 for Windows (SPSS, Chicago, IL, USA). Results were expressed as mean ± SD. Statistical significance was defined as P < 0.05. Categorical variables were analysed by the χ² test and continuous variables with the paired or unpaired Student's t-test. Life table analysis was used to calculate complete sternal fusion rates and the log-rank test was utilized to test for differences.

RESULTS

Sternal pins were utilized in 36 patients for sternal reinforcement in addition to the standard closure of the sternum with Ethibond sutures (Group A), and 27 received no pins with the standard Ethibond sternal closure (Group B).

Pre- and intraoperative patient profiles are described in Table 1. There were no significant differences between the groups in terms of preoperative characteristics. In Group A, 34 of 36 patients underwent ascending aorta and total aortic arch replacements, and two underwent ascending aorta and hemi-aortic arch replacements. In Group B, 24 of 27 patients underwent ascending aorta and total aortic arch replacements, and three underwent ascending aorta and hemi-aortic arch replacements.

Table 1:

Pre- and intraoperative patient profiles

	Group A	Group B	P-value
No. of patients	36	27
Age	66.4 ± 10.5	67.7 ± 8.97	NS
Male (%)	29 (80.6)	18 (66.7)	NS
Diabetes mellitus (%)	6 (16.7)	4 (14.8)	NS
Chronic renal failure (%)	4 (11.1)	3 (11.1)	NS
Chronic lung diseases (%)	5 (13.8)	3 (11.1)	NS
Peripheral vascular diseases (%)	2 (5.6)	2 (7.4)	NS
Stroke (%)	8 (22.2)	4 (14.8)	NS
Emergency (%)	3 (8.4)	3 (11.1)	NS
Procedures
Ascending aorta and total aortic arch replacement (%)	34 (94.4)	24 (88.9)	NS
Ascending aorta and hemiarch replacement (%)	2 (5.6)	3 (11.1)
Operative time (min)	530 ± 134	561 ± 275	NS
Cardiopulmonary bypass time (min)	254 ± 55	269 ± 72	NS
Concomitant procedures
Aortic root surgery	2	2	NS
AVR	3	2	NS
CABG	2	2	NS

AVR: Aortic valve replacement; CABG: coronary artery bypass grafting.

Clinical results

There were no perioperative deaths. Two patients were transferred back to the operating room for postoperative bleeding.

The mean follow-up term was 7.2 ± 7.2 months (1–24 months). Follow-up was completed in all patients. Neither evident sternal infection nor sternal instability which required surgical intervention was observed in either group during the follow-up period.

Computed tomographic evaluations

Sternal dehiscence

Two sternal dehiscences (5.6%) in Group A and four dehiscences (14.8%) in Group B were observed on the discharge CT scans. No significant difference was found between the groups in terms of the occurrence of the early postoperative sternal dehiscence (P = 0.215). During the follow-up, another two dehiscences occurred in Group B. No late sternal dehiscence was observed in Group A (Table 2).

Table 2:

Incidence of sternal dehiscence, displacement and fusion

	Dehiscence		Displacement		Fusion
	Group A	Group B	Group A	Group B	Group A	Group B
At discharge (%)	2 (5.6)	4 (14.8)#	0 (0.0)	5 (18.5)**	0 (0.0)	0 (0.0)#
Late follow-up (%)	0 (0.0)	2 (7.4)	0 (0.0)	2 (7.4)	14 (38.9)	7** (25.9)
Total (%)	2 (5.6)	6 (22.2)*	0 (0.0)	7 (25.9)**	14 (38.9)	7** (25.9)

^#Not significant versus Group A.

*P < 0.05 versus Group A.

**P < 0.01 versus Group A.

Sternal displacement

There was no occurrence of anteroposterior sternal displacement (0%) in Group A at discharge. Meanwhile, five displacements (18.5%) were observed in Group B at discharge. There was a significant difference between the groups in terms of the occurrence of early sternal displacement (P = 0.007). During the follow-up, two more displacements were observed in Group B. No late sternal displacement occurred in Group A (Table 2).

Sternal fusion

Fourteen patients (38.9%) achieved complete sternal fusion in Group A during the study period. In contrast, only seven patients (25.9%) in Group B could achieve complete sternal fusion. The complete sternal fusion rates at 12 months postoperatively were 100% in Group A and 21.6% in Group B (P < 0.001) (Fig. 2).

Figure 2:

Comparison of the cumulative complete sternal fusion rate between the Groups A and B.

Time-dependent changes of cortical bone density area

Figure 3 shows time-dependent changes of the CBDA in both groups. A full 12-month follow-up was obtained from 14 patients in Group A and from 11 patients in Group B. In Group A, the mean CBDA was 332.9 ± 99.3 mm², 400.2 ± 87.0 mm² (120.2% of the discharge value), 368.1 ± 98.4 mm² (110.6%) and 431.1 ± 163.4 mm² (129.5%), at discharge, 3, 6 and 12 months post-operative, respectively. There was a significant increase in the CBDA in Group A (P < 0.001; between at discharge and 12 months). In Group B, the mean CBDA was 285.9 ± 169.4, 233.4 ± 61.32 mm² (81.6%), 314.5 ± 167.5 mm² (110.0%) and 318.4 ± 70.8 mm² (111.4%), at discharge, 3, 6 and 12 months post-operative, respectively (not significant; between at discharge and 12 months). It is impressive that osteoconduction seems to occur around the sternal pins, and the anterior and posterior plates of the sternum were strongly reinforced internally by the growing CBDA (Fig. 4).

Figure 3:

Time-dependent changes of the cross-sectional CBDA. The value at discharge was expressed as 100% for every patient. The value at 12 months postoperatively was divided by the value at discharge, and expressed as a percentage compared with the value at discharge.

Figure 4:

CT images of the sample slice at discharge and 12 months post-operatively in Group A. The remarkable increase of the cross-sectional CBDA was observed. The anterior and posterior plates of the sternum were strongly reinforced internally by the growing CBDA.

DISCUSSION

Sternal instability after a sternotomy can be responsible for delayed wound healing or the development of sternal infections [1]. A bioresorbable radiopaque osteosynthesis device, (Super Fixsorb 30), a mixture of PLLA and u-HA, is considered to have a high mechanical strength, total resorbability and osteoconductivity [8]. Many kinds of screws, pins, plates and other internal bone fixation devices made of this material are now in clinical use for orthopaedic, oral and maxillofacial, craniofacial and plastic reconstructive surgeries [9–11]. The Super Fixorb 30 sternal pin was approved by the Conformité Européenne as a new generation bone-bonding material in the cardiac and thoracic regions (Osteotrans-OT Rib/Sternum Pin), and there have been several reports of cardiothoracic surgeries using the material for sternal reinforcement or rib fixation with favourable results [12, 13].

In the current study, we found that the sternal pin could reduce the incidence of sternal displacement in an anterior–posterior direction. Most of the displacements are believed to be due to inappropriate sternal approximation. Ethibond sutures or steel wires can prevent the two pieces of the sternum from separating, but not from slipping in an anterior–posterior direction. The sternal pins can prevent the two pieces of the sternum from slipping by anchoring a piece of sternum to the other one, and the ideal sternal approximation can be obtained. Our results demonstrated a significantly high sternal fusion rate in the patients with the sternal pins. The adequate approximation of the sternum in an anterior–posterior direction may have facilitated earlier sternal fusion in patients with the sternal pins.

Despite the high dehiscence rate in both groups, we fortunately did not experience either sternal instability or infection. We presumed that most of the dehiscences were spatially reinforced by the connective tissue ingrowth, which is frequently observed in redo sternotomy. The tissue ingrowth may contribute to preventing sternal instability or infection. However, we consider that this type of healing is still inappropriate, and should be avoided to prevent future complications.

Super Fixsorb 30, a mixture of u-HA and PLLA, was reported to have a better osteoconductivity than PLLA-only devices, especially when they were placed in direct contact with the bone [8]. We investigated the osteoconductivity of this material by evaluating the time-dependent changes of the cross-sectional CBDA of the sternum. The results showed the time-dependent increase of the CBDA around the sternal pins. It is particularly worth noting that the bone marrow space around the pins had changed into the CBDA, and the anterior and posterior sternal plates had increased their thickness. This result suggests that the sternal pin can have a potential of inducing a bone conduction within itself as well as an osteogenesis around it. This effect can also result in favourable influences on earlier sternal fusion in patients with the sternal pins.

Little information is available in terms of the bioresorbability of the material. Shikinami et al. [8] reported that 4–5 years was required to absorb the composite pin which was placed into the femur of a rabbit, and that particles of hydroxyapatite were still microscopically observed.

In terms of clinical impacts, Hamaji et al. [12] reported favourable results of the reinforced sternal closure using the same sternal pin with an extremely low sternal infection rate even in a high-risk cohort. The Super Fixsorb 30 is admittedly a foreign material for a living body. Therefore, it has a potential effect of inducing inflammatory or infectious reactions. In the current study, we observed no deep sternal infections. This result supports that the material will not trigger adverse infectious reactions after a sternal closure. The bioresorbability of this material can be one of the advantages for preventing late inflammation or infection. However, we admit that the number of the patients in the study is too small to conclude that the pin could contribute to reducing the risk of sternal infections. Large numbered, randomized or case-matched studies should be conducted to prove the clinical advantage of the material for preventing sternal infections.

Limitations

This study is a single institutional clinical review, non-randomized, non-matched and retrospective in nature, which is subject to the inherent weaknesses of a retrospective analysis. Another weakness of the present study is the relatively small number of patients, which may have been underpowered to detect subtle statistical significances, especially in analysing the impacts of the pin on the clinical outcomes.

In conclusion, the use of the Super Fixsorb 30 sternal pin reduced anterior–posterior sternal displacement and facilitated earlier sternal fusion. The sternal pin may promote osteogenesis of the sternal bone around it and contribute to earlier sternal fusion.

Supplementary material

Supplementary material is available at ICVTS online.

Conflict of interest: none declared.