Summary
Background
Open-heart surgery can result in adhesions, which can complicate resternotomy.
Aims
To document the occurrence of adhesions after the use of a new collagen membrane; to evaluate its tolerability; and to compare surgical parameters with control patients.
Methods
Paediatric patients who underwent cardiac surgery with the collagen membrane (Cova™ CARD; Biom’up, Saint Priest, France) were analysed retrospectively for levels of adhesion and tolerability. The times of dissection and intervention and the transfusion of packed red blood cells and haemostatic products were compared to a historic cohort who did not receive an anti-adhesion device.
Results
From January 2010 to December 2011, 36 patients received a collagen membrane. Nineteen re-interventions were performed, after a mean of 169 days. No grade 3 adhesions were observed and no tolerability problems were reported. During re-interventions after more than 30 days, the propensity score-adjusted durations of dissection and the total process for patients with and without a collagen membrane were 32 vs 41 minutes and 151 vs 182 minutes, respectively (not significant). The mean quantities of red blood cells and biological glue administered in the two groups were 98 vs 139 mL and 1.2 vs 0.5 mL, respectively (not significant).
Conclusions
This feasibility study shows the potential use of the new membrane in paediatric patients, both in terms of prevention from severe adherence and tolerability. This is the first study of this membrane in humans. A prospective, controlled study is necessary to provide strong evidence of its efficiency.
Résumé
Contexte
La réalisation de chirurgies à cœur ouvert peut entraîner des adhérences qui peuvent compliquer la re-sternotomie.
Objectif
Évaluer la survenue d’adhérences après utilisation d’une membrane en collagène ; évaluer sa tolérance dans une population pédiatrique ; et comparer des paramètres chirurgicaux chez les témoins.
Méthodes
Une étude rétrospective a été conduite chez des enfants ayant subi une chirurgie cardiaque et ayant bénéficié de la membrane de collagène (Cova™ CARD ; Biom’up, Saint Priest, France). Les degrés d’adhérence à la réintervention chirurgicale et les données de tolérance ont été colligés. Les durées de dissection chirurgicale et réintervention, l’utilisation de culots globulaires rouges et d’agents hémostatiques ont été comparées à une série de patients n’ayant pas bénéficié d’anti-adhérent.
Résultats
Entre janvier 2010 et décembre 2011, 36 patients ont été sélectionnés dans le groupe membrane. Dix-neuf réinterventions ont été réalisées après un délai moyen de 169 jours. Aucune adhésion de grade sévère n’a été observée ni de problème de tolérance. Pour les réinterventions après 30 jours, les durées de dissection et d’intervention ont été respectivement de 32 et 151 minutes contre 41 et 182 minutes dans le groupe témoin (non significatif). Les quantités moyennes de culots transfusés et de colles biologiques utilisées dans les groupes ont été respectivement de 98 contre 139 mL et de 1,2 contre 0,5 mL (non significatif).
Conclusions
Cette étude de faisabilité montre l’intérêt potentiel de la nouvelle membrane dans une population pédiatrique tant sur le plan de la prévention des adhérences sévères que de la tolérance. C’est la première étude chez l’homme avec cette membrane. Une étude prospective contrôlée est nécessaire pour apporter la preuve de son éventuelle supériorité par rapport aux stratégies habituelles.
Background
Cardiac reoperations are increasingly common, especially among paediatric patients with cardiac disease. Approximately 10% of cardiac operations involve repeat sternotomies . There is a risk of re-entry injury due to the presence of adhesions that can obscure the cardiac anatomy . The surgical trauma to the pericardiac mesothelium during open-heart surgery can result in a decrease in fibrinolytic activity and the formation of adhesions at the intervention sites . Therefore, re-interventions may be complicated by the development of dense, fibrous retrosternal or pericardiac adhesions. As a consequence, re-sternotomies can be dangerous and surgery is complicated because the surgeon needs to use sharp instruments that might damage the tissues and cause potentially serious iatrogenic haemorrhagic events . Adverse events related to reoperation can lead to poor patient outcome and higher cost .
These complications of reoperations have prompted experimental studies to prevent or decrease adhesion formation and thus decrease the risks associated with reoperation.
Here, we describe our clinical experience of the use of a new collagen membrane which aims to prevent surgical adhesions.
Methods
Study type
This single-centre, retrospective cohort study evaluated a group of patients treated with a collagen membrane, and compared this group to a historic cohort of patients not treated with a membrane.
Objectives
Our main objectives were to evaluate: (1) the occurrence of adhesions during surgical re-interventions with collagen membrane and (2) the tolerability and the morbidity related to the presence of adhesions after the use of the collagen membrane. Our secondary objectives were to evaluate and compare the duration of dissection, the total duration of re-intervention and the volumes of packed red blood cells (RBC) and biological glue used during surgical re-interventions among those treated with and without collagen membrane.
Evaluation criteria
As there is no specific classification for the severity of adhesions in cardiopaediatric surgery, adhesions among patients in the collagen membrane group were graded according to the levels of adhesion described by Lodge et al. ( Table 1 ). No record of adhesion severity in the control arm was available, as we only started using this classification scheme after the introduction of the collagen membrane.
| Description | |
|---|---|
| Level of adhesion | |
| Grade 0 | Absence of adhesion |
| Grade 1 | Low adhesion (possibility to lyse it with the finger) |
| Grade 2 | Moderate adhesion (possibility to lyse it with scissors) |
| Grade 3 | High adhesion requiring the use of an electrical bistoury in order to perform the dissection |
| Adhesion risk zone | |
| A | Anatomic structures located in the sternotomy zone |
| B | Necessary mediastinum elements for the installation of ECC (ascending aorta, superior vena cava, part of the right atrium) |
Adhesions among collagen membrane recipients were also characterized by the risk zone of their occurrence, as described in Table 1 . Zone A is the first zone encountered during re-intervention. Resternotomy in a patient with adhesions in zone A is accompanied by the risks of damaging the heart or any other structure that can stick to the sternum. This can be fatal, as the ECC has not yet been implemented, and the patient may die from haemorrhagic shock, especially as access to a femoro–femoral ECC is impossible in young children. Adhesions in zone B can affect the second stage of surgery, which begins once the sternal retractor is in place and the structures needed to start the ECC (aorta and right atrium) are dissected. This step is somewhat less dangerous because the wound can be more easily controlled as the heart is better exposed. The morbidity related to the presence of adhesions was assessed as any complication related to the presence of adhesions mentioned in the medical charts, regardless of the severity.
The evaluation of adhesion grade was not blinded. Potential undesirable effects related to the collagen membrane, such as inflammatory or allergic reactions, were collected in order to evaluate its tolerability.
As adhesions were not recorded in the control arm, the following objective parameters were retrieved for comparison between the groups. The duration of dissection of the re-intervention was the time between the first surgical incision and the start of ECC. The volume of RBC was assessed, taking into account all administration of RBC during the procedure. The volume of biological glue used during re-intervention was the total amount of Tissucol ® (Baxter, Deerfield, USA).
Study membrane and procedure
The study membrane is a resorbable, malleable porcine collagen membrane called Cova™ CARD (Biom’up, Saint Priest, France) ( Fig. 1 ), which acts as a barrier between the heart and the surrounding tissues. It also promotes tissue regeneration . It is a class III device that obtained a CE approval in 2009. The membrane is available in two sizes (30 × 40 mm and 40 × 60 mm), according to the weight of the child. The membrane has to be moisturized for a few minutes in a physiological serum to make it more flexible. It can be cut as necessary, in order to be able to place it on delicate structures; and can be fixed with sutures. It is applied next to the right atrium, the aorta and the sternum, before the sternal closure.
Selection criteria for the collagen membrane group
Patients in the collagen membrane group were those who had received the collagen membrane during January 2010 to December 2011 during surgery for a congenital cardiopathy requiring repeat scheduled open-heart surgery with a first re-intervention scheduled less than 24 months after the first intervention with median sternotomy and requiring ECC.
Selection criteria for the control population
Control patients were those who had an initial intervention using an approach with median sternotomy during 2009; had a re-intervention performed by the same cardiac surgeon as the re-interventions performed in the membrane group; had a re-intervention occurring more than 30 days after the original procedure that was performed using an approach with median sternotomy and requiring ECC. In all control patients, the pericardium was not closed after the first procedure.
Data collection and analysis
Baseline parameters are presented as means and standard deviations (SDs) for quantitative variables and numbers (percentages) for qualitative variables. To adjust for baseline characteristics between the two groups, we estimated propensity scores using a logistic regression model (gender, age at initial surgery, delay of re-intervention, age at re-intervention and weight at revision surgery). We then compared re-intervention parameters using linear models with a covariate adjustment on these propensity scores. Adjusted means with 95% confidence intervals (CIs) are presented.
Results
Use of the collagen membrane
From January 2010 to December 2011, 36 children (25 boys; 11 girls) with congenital cardiopathy received a collagen membrane after a first cardiac surgery and were included in this study. Among them, two received a second membrane after a second surgery and one patient received a total of four membranes after four surgeries performed during the observation period. In total, 41 collagen membranes were implanted during 41 surgeries. The mean age of the patients at the time of the first intervention was 13.8 ± 33.7 months (range: 0.1–200 months).
Nineteen re-interventions were undertaken during the observation period on a total of 14 patients, of whom two had two re-interventions and one had four re-interventions. No re-intervention was due to an undesirable event related to the initially implanted anti-adhesion membrane. Re-interventions occurred within a mean delay after initial surgery of 169 ± 166 days (range: 3–558 days). Twelve re-interventions (63%) occurred after a delay of more than 30 days.
The adhesion levels for the 19 re-interventions performed after the installation of the collagen membranes are shown in Fig. 2 . In risk zone A (sternotomy zone), regardless of the re-intervention delay, no grade 2/3 adhesion was observed and 89% of patients had no adhesion ( Fig. 2 A). Beyond a re-intervention delay of 30 days, 17% of patients had grade 1 adhesions. In risk zone B (ECC zone), there were some grade 1/2 adhesions, but no severe adhesions (grade 3) ( Fig. 2 B). Beyond a re-intervention delay of 30 days, the rates of grade 1 and grade 2 adhesions were 42% and 25%, respectively. Tolerability of the membrane was good as no serious or minor undesirable effects were observed, in terms of infection, allergy, intolerance, inflammatory signs or secondary vascular rupture. There were no complications due to the presence of adhesions during the re-interventions performed in the membrane group; and no haemorrhagic accidents were recorded.
