| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
Interact CardioVasc Thorac Surg 2009;9:565-570. doi:10.1510/icvts.2008.196709 © 2009 European Association of Cardio-Thoracic Surgery
Growth potential of U-clipTM interrupted versus polypropylene running suture anastomosis in congenital cardiac surgery: intermediate term results
a Clinic for Cardiovascular Surgery, Swiss Cardiovascular Center Bern, University Hospital, Bern, Switzerland Received 22 October 2008; received in revised form 30 June 2009; accepted 6 July 2009
*Corresponding author. Klinik Im Park, Seestrasse 220, Zurich, 8027, Switzerland. Tel.: +41 44 2092552; fax: +41 44 2092512.
Although U-clip anastomoses were studied for hemodynamics and patency, their potential for unimpeded growth after congenital cardiovascular surgery has not been investigated yet. In 53 children aged 2.1±3.3 years operated on between March 1998 and August 2005 growth of U-clip (U) vs. polypropylene running sutured (P) anastomoses in coarctation repair (Coarc; n=26), bi-directional Glenn (BDG; n=13) and arterial switch operation (ASO; n=14) was retrospectively analysed. Coarc showed 2.39±4.33 vs. 3.09±2.24 mm of growth during the observation period (21±16 vs. 30±27 months); no growth (0 vs.16%), restenosis (14 vs. 37%) and reinterventions (14 vs. 11%) were similar (all in U vs. P, P=ns). BDG showed 3.68±3.43 vs. 2.50±2.55 mm (P=ns) of growth during 15±5 vs. 29±18 months (P=0.046); no growth (17 vs. 0%), stenosis (0 vs. 14%) and reinterventions (0%) were similar in U vs. P, respectively (P=ns). Main pulmonary artery (MPA) anastomosis in ASO showed 0.28±1.73 vs. 1.30±3.16 mm of growth during 8±14 vs. 28±28 months; no growth (60 vs. 14%), stenosis (50 vs. 63%) and reinterventions (0%) were similar (all in U vs. P, P=ns). Anastomotic growth, stenosis and reintervention rates show no difference between interrupted U-clip and polypropylene running sutured technique in Coarc repair, BDG and MPA anastomosis in ASO.
Key Words: Cardiac surgery; Congenital; Vascular; Growth
Construction of vascular anastomoses with interrupted sutures demands considerable skill and may be time-consuming. Nevertheless, interrupted sutures avoid purse-string effects, maximize anastomotic diameter, pulsatility and unrestricted growth in comparison to running sutures [1–3]. They may, therefore, be preferable, especially with microvascular anastomoses, low-pressure vessels, and during growth [4, 5]. However, because of the time-consuming process, extensive manipulation and knot-tying most surgeons still perform running sutures. With the U-clipTM (U) (Medtronic Inc, Minneapolis, MN, USA), a nitinol self-coiling penetrating clip, construction of interrupted anastomoses is strongly facilitated (Fig. 1a, b). U-clip anastomoses have demonstrated excellent patency, healing characteristics and significantly greater compliance and cross-sectional area compared to running sutured ones [6–8]. Because of its potential advantages regarding growth we have started to use the U-clips also in children [9].
U-clips were used in three surgical procedures – coarctation repair (Coarc), bi-directional Glenn anastomosis (BDG) and arterial switch operation (ASO) – because restenosis is a well-known complication both after Coarc in up to 15% [10] and on the main pulmonary artery (MPA) after ASO in 2.3–9.2% [11–13]. U-clips were also thought to ensure unimpeded growth and maximize the size of the low-pressure BDG anastomosis. The purpose of this study is, therefore, to determine the growth pattern of anastomoses with U-clips (U) compared to polypropylene running sutures (P) in pediatric cardiac surgery. Furthermore, this study also provides comparison of the growth of anastomosed to normal vessels.
Fifty-three pediatric patients operated on at our institution from March 1998 to August 2005 with a mean age of 2.1±3.3 years (4 days–13.7 years) for Coarc (n=26), BDG (n=14) and ASO (n=14) were included in this retrospective cohort study; eight patients were excluded for lacking data. Patients' baseline characteristics were not different between U and P (Table 1).
Primary endpoints of this study were absence of anastomotic growth, reoperation or reintervention for (re-) stenosis and death from any cause. The U-clip technique is described in detail elsewhere [9]. In Coarc, U-clips were used for the whole anastomosis (n=5) or for at least the anterior two-thirds of the aortic circumference, while the dorsal wall was sutured with a polypropylene running suture (n=2). In BDG, the anastomosis is performed using U-clips for the whole (n=5) or in one case for the anterior two-thirds of the circumference. In ASO, the MPA is reconstructed with a xenopericardial pantaloon patch, contributing to approximately two-thirds of the distal anastomotic circumference. The residual anterior one-third of circumference, at which native vessel tissue joins from both ends and therefore, constitutes the only growth zone, is re-anastomosed with U-clips. In all available preoperative and follow-up echocardiographic and catheter examinations anastomotic width was analyzed. Patients were stratified by anastomotic technique in two cohorts (U and P) within the three surgical groups. Patients' selection to group U or P was basically historical, since patients operated on before 2001 had polypropylene running sutured anastomoses and those operated on later had, in general, U-clip sutures. 2.2. Measurement techniques and definition of growth All diameters were manually measured in echocardiograms, selecting the most appropriate images, and the angiograms were analysed using Philips Inturis Suite V 1.1.2.In Coarc, width of the ascending aorta, the mid aortic arch and the aorta at the anastomosis was assessed. In BDG, anastomotic superior vena cava (SVC) diameter was measured in a–p exposure at angiography. In ASO, MPA width at the anastomosis was measured. All examinations had been performed by the same two cardiologists (JP and MP), and all post-hoc measurements were done by the same investigator (JL). Due to the estimated variance in manual post-hoc measurements, we defined positive and negative growth as a difference in diameter exceeding ±1 mm. For follow-up periods of over 600 days we accepted every difference in diameter as positive or negative growth. ‘No growth’ was defined as a difference in diameter below 1 mm. Relative growth was defined as increase in diameter/body surface area (BSA). Data are expressed as mean value±standard deviation (S.D.). Ninety-five percent confidence intervals (CI) and percentages are given where appropriate. Statistical analysis was performed using Statview 5.0.1 for Windows (SAS Institute Inc, Cary, NC, USA). For univariate analysis, the Mann–Whitney U-test for continuous data and the Fisher's exact test for contingency tables for nominal data were used. A P<0.05 was determined statistically significant.
Adequate somatic growth was seen in all groups showing no significant difference between U and P in all groups. Changes in BSA over time paralleled within all groups between U and P. BSA of most patients was within the range of the 3rd and 97th percentiles of their age-matched peers during follow-up. Coarc anastomosis showed an overall growth of 2.39±4.33 (95% CI –1.62–6.40) mm in U and 3.09±2.24 (95% CI 1.62–3.90) mm in P during an overall follow-up period of 21±16 and 30±27 months in U and P, respectively (all P=ns; Fig. 2a). Increase in anastomotic diameter paralleled between U and P and showed a late decrease in both subgroups. No growth was observed in 0% (0/7 patients) and 16% (3/19 patients) with 1 (14%) vs. 7 (37%) restenoses and 1 (14%) vs. 2 (11%) reinterventions with U and P, respectively (all P=ns). Compared to normal diameters of the respective aortic segments, distal ascending aorta with both U and P remained smaller (P=0.07), whereas at the anastomotic level the aorta was larger than normal with U, but smaller than normal with P (P=0.39) (Fig. 2b, c) at 30 days postoperatively. Relative growth of the anastomotic segment of the aorta is overall identical between U and P (Fig. 3a). Compared to normal values, relative growth with U is identical within 30 days postoperatively, whereas with P relative growth is slower than U and normal (Fig. 3a, b). Within U, no differences were seen between anastomoses completely made with U compared to those made with U and running polypropylene suture.
3.2. BDG anastomosis BDG anastomosis showed an overall growth of 3.68±3.43 (95% CI 0.07–8.56) mm in U and 2.50±2.55 (95% CI 1.05–4.40) mm in P (P=ns) during a follow-up of 15±5 and 29±18 months in U and P, respectively (P=0.046). Whereas, over time, BDG anastomosis constantly grew in U, it remained stable in P (Fig. 4a). No growth was observed in 17% (1/6 patients) and 0% (0/7 patients) with 0 vs. 1 (14%) restenosis and no reintervention in U and P, respectively (all P=ns). Compared to normal SVC diameter of 11.5 mm in patients with 10 kg body weight and 0.5 m2 BSA, initial BDG diameters in our study of 11 and 13 mm in subgroups P and U were larger, taking into account the smaller weight (6.9±1.4 kg) and BSA (0.35±0.05 m2) of subgroup P patients. Normalizing BDG diameter to BSA shows a relative anastomotic increase in U and a decrease in P (Fig. 4b). Within U, no differences were seen between anastomoses completely made with U-clips compared to the one made with U-clips and running polypropylene suture.
3.3. ASO MPA anastomosis after ASO showed an overall growth of 0.28±1.73 (95% CI –1.87–2.43) mm in U and 1.30±3.16 (95% CI –1.20–3.70) mm in P (P=ns) during a follow-up of 8±14 and 28±28 months in U and P, respectively (P=0.07) (Fig. 5a). No growth was observed in 60% (3/5 patients) and 14% (1/7 patients) with 3 (50%) vs. 5 (63%) restenoses and no reinterventions in U and P, respectively (all P=ns).
In the first postoperative year, MPA diameters remained unchanged in both U and P and increased thereafter in both subgroups. MPA in the first year of life increased from 8.7 to 12.3 mm for a normal child with comparable BSA of 0.2 and 0.4 m2 as in our ASO group. MPA diameters after ASO seen in our study were therefore almost normalized with surgery in both subgroups (8.4±1.5 and 7.3±1.9 mm), however, they failed to grow normally within the first year postoperatively (9.3±3.9 and 9.0 mm). Relative growth of MPA after ASO is comparable to normal after U. Normal MPA diameters, however, remain larger compared to the first 100 days postoperatively (Fig. 5b, c). With P, MPA diameters are comparable to normal values after 30 days. Relative growth, however, slows considerably, so that by 100 days postoperatively, relative diameters are much smaller than U and normal (Fig. 5b, c). Restenosis, re-intervention, reoperation and death rates are summarized in Table 2. No statistically significant difference was observed neither for a single nor a combined clinical endpoint (all four endpoints cumulated), calculated for number of events and of patients. The four deaths occurring in this series were not related to the anastomotic technique used.
Although in children, anastomotic stricture or restenosis may be due to many different factors, such as inadequate tissue growth, excessive scar formation, use of foreign material or myointimal hyperplasia, it may also be propagated by suture materials and techniques used, and may make re-interventions or reoperations necessary. Studies performed in the 1980s and 1990s demonstrated advantages of interrupted over running sutures concerning their potential to preserve vascular growth in porcine and canine models [1–3]. U-clips facilitate precise and rapid creation of an interrupted anastomosis, appearing especially useful in preserving growth potential superior to running sutures [9]. However, we could not confirm this hypothesis. In contrast to the literature, where interrupted suture techniques proved superior to running ones [1–3], interrupted U-clips suturing in our study showed not to be superior to conventional non-absorbable running sutures concerning anastomotic growth or restenosis in the clinical setting of Coarc, Glenn anastomosis and MPA reconstruction in ASO. In all three groups qualitative as well as quantitative assessment demonstrated comparable growth rates for both suturing techniques. Still, some differences, albeit temporarily seen and statistically not significant, deserve some comment. A larger than normal anastomotic segment at the level of the proximal descending aorta after U but not P, although not significantly, may be the result of prevention of some purse-string effect with U and the fact, that with the use of U or another interrupted technique, anastomotic cross-sectional area is increased up to 150% [2, 7]. Another positive effect of U is seen with BDG. Despite a longer median follow-up of 10 months in BDG with P, anastomotic diameters remained almost unchanged in contrast to BDG with U, where diameters constantly increased during a shorter follow-up period. This effect was still seen after normalisation for BSA, indicating increased anastomotic relative to somatic growth. This observation may indicate an especially beneficial effect of U anastomoses in low-pressure vessels. Comparison to normal MPA diameter shows, that the MPA is enlarged to normal diameter during ASO. But it then fails to grow with both techniques in the first postoperative year in most patients (60% and 14%), growing only later and leading to stenosis in a high proportion of patients (50% and 62.5%). This poor growth observed may be due to the rather short follow-up duration and may well be compensated later. The U-clip technique alone may, therefore, not be effective enough to reduce the risk for supravalvular pulmonary stenosis, the most frequent cause of reoperation in 2.3–9.2% after ASO [11–13]. Despite the use of the pantaloon patch technique for MPA reconstruction, known to provide superior results compared to the button technique [12], poor growth is seen in our series. This delayed and inadequate growth may be due to various factors. First, the one-third of circumference of tissue-to-tissue anastomosis remaining as the growth zone may not have enough intrinsic potential to provide adequate growth, even when interrupted sutures are used. Second, scarring at the anastomosis as well as around the vessel, which may be accentuated by the presence of patch material, may prevent the MPA from growing adequately. Several limitations may have influenced the findings of this study. There are the limitations inherent to a retrospective trial. However, with a matched control group important baseline characteristics were controlled. Since the groups were rather small, the study may be underpowered to identify subtle differences. The observation period may be too short to draw definitive conclusions. However, this may only be true for ASO with U with a relatively short median observation period of eight months. In the literature, supravalvar PS mainly occurs within the first two years postoperatively with a second risk hazard phase for re-intervention after 4–7 years and a constant reoperation rate of 1% per year [11–13]. Too short an observation period may not apply for the other two groups, since follow-up period covered the time of most frequent occurrence of restenosis 2–3 years postoperatively. In conclusion, anastomotic growth shows no difference between U-clip and polypropylene running sutured technique in coarctation repair, Glenn anastomosis and MPA reconstruction during ASO. Restenosis, re-intervention or reoperation rates remained unaffected by the anastomotic technique used. Larger groups and longer observation periods are needed to demonstrate superior growth preservation and less restenosis of one or the other anastomotic technique in the future.
We thank Brigitta Gahl, statistician for her support and advice on statistical analysis.
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
Top
Abstract
1. Introduction
