Abstract
BACKGROUND:
There has been increased interest in the use of biomaterials that resorb completely leaving only the patient’s native tissue. Synthetic materials are advantageous for tissue repair because they are highly customisable. The infection rate of using resorbable natural materials in paediatric surgery has recently been outlined, but there has not yet been a review of the use of synthetic resorbable materials in paediatric surgery.
OBJECTIVES:
This systematic review analyses the risk of infection after implantation of fully resorbable synthetic biomaterials in paediatric cases.
METHODS:
The literature was searched from January 1970 to January 2018 (inclusive), specifically searching for paediatric cases (0–18 years old), use of synthetic resorbable materials and infection.
RESULTS:
The infection rate in 3573 cases of synthetic resorbable material implantation was 1.1% (41 cases). A Chi-squared test for independence found infection rate to vary among materials. Of the many biomaterials identified in this review, the highest infection rates were seen in Suprathel’s use in burns injuries (12.1%).
CONCLUSIONS:
This review found a low infection rate in synthetic resorbable materials used in paediatric surgery, with particularly strong evidence for low infection risk in LactoSorb® use.
Background
Saxena et al. recently outlined the infection rate of using resorbable natural biomaterials (Integra, Surgisis, Pelnac, Permacol, Tutopatch and Lyodura) in paediatric surgery [1]. These materials are made of the components of extracellular matrix; for example, different types of collagen and glycosaminoglycans. The authors reported the infection rate as 5% across 298 paediatric cases included in the review. Burn wound cases had a higher infection rate than non-burn wound cases, and infection rate varied among materials used (for example, Pelnac had a higher perioperative infection rate). These natural materials have the advantage of being bioactive and able to mimic native tissue, encouraging the healing process in the same way that the host environment does. Another consequence of mimicking native tissue is their low toxicity profile. On the other hand, natural materials have the disadvantage of poor customisability, and are often expensive to manufacture. Their bioactivity can also be a source of immunogenicity to the host. Saxena et al.’s review offered further insight into the risks of using these materials in the paediatric population.
No review to date has examined the risk of infection in synthetic resorbable material implantation in paediatric surgery. Synthetic materials are advantageous for their highly customisable structural, chemical and degradation properties. They are also cheaper to manufacture. However, they carry the risk of toxicity and immunogenicity to the host due to their significant difference to native tissue [2]. This review aims to determine the infection rate when using resorbable synthetic biomaterials in paediatric surgery, clarifying their safety profile.
Methods
The literature was searched between January 1970 and January 2018 (inclusive) in MEDLINE and EMBASE. Only publications on paediatric patients (0–18 years old) undergoing repair with resorbable materials were included. Individual case reports were excluded, as that study design was considered a potential source of bias. Articles retrieved from that search were assessed for inclusion in this review according to the PRISMA flow-chart (Fig. 1).
As demonstrated in the PRISMA flow chart, 46 articles were excluded based on assessment of the full text; 14 of these articles were eventually deemed irrelevant, 9 were reviews of inappropriate materials for this review, 8 were not available as English full texts, 5 articles reported use of multiple synthetic materials in their study but did not detail the exact number of patients receiving each material, 5 articles could only be found in abstract form, 4 articles either only reviewed adult cases or did not differentiate between paediatric and adult cases, and 1 article reviewed paediatric cases of resorbable material use but did not specify infection as an outcome.
PRISMA flow chart of MEDLINE and EMBASE literature search.
A Chi-squared test for independence was performed to statistically compare the infection rates among materials. A p value of <0.05 was considered to be statistically significant.
Eleven synthetic resorbable materials for use in paediatric tissue repair were found. These materials were used in a total of 3573 surgical procedures on paediatric patients, of whom 41 were reported to have developed an infection post-operatively (infection rate = 1.1%). Particular attention was paid to the follow-up times of the studies, as infection may occur at any time before resorption, even if infection is more likely to occur early after implantation due to perioperative contamination [3,4]. The vast majority of literature identified focused on craniofacial surgery. The composition of each material is displayed (Table 1). Infection rates for each individual material are included below, and in Table 2.
Composition of synthetic resorbable biomaterials used for paediatric surgery
Composition of synthetic resorbable biomaterials used for paediatric surgery
Infection rates of synthetic resorbable biomaterials used for paediatric surgery
LactoSorb® is a widely used synthetic resorbable biomaterial for paediatric tissue repair. It is a blend of 82% poly-l-lactic acid (PLLA) and 18% polyglycolic acid (PGA), and fully resorbs within 6–12 months. Ahmad et al. performed a prospectively designed study on LactoSorb® use in 146 patients who underwent surgery for craniosynostosis, although 4 patients were lost to follow-up. The authors followed up outcomes in all patients at 3, 6, 9 and 12 months, and 5 out of 142 patients developed infection post-operatively [4]. Another prospectively designed study by Eppley et al., with clearly defined patient inclusion criteria and follow-up times (85 patients reviewed at 12 months, encompassing the time for total resorption, and 15 reviewed after 6 months), found no infections in 100 patients who underwent craniofacial surgery with LactoSorb® implants [5]. Eppley et al. later performed a combined prospective and retrospective review of 1883 patients who also underwent craniofacial surgery with LactoSorb® plate implantation, and found 8 patients developed infection, 3 of whom needed hospital readmission to resolve the infection [6]. Several more retrospective studies found similar results. Munoz-Casado et al. reviewed 216 cases of craniosynostosis repair; 199 of which used Lactosorb® (inferred from percentage values found in full-text) and didn’t report any case of infection [7]. Branch et al. reviewed 203 cases of craniosynostosis repair, 150 of which used LactoSorb® (inferred from percentage) and found 3 cases of infection [8]. Burstein et al. reviewed 60 cases of frontal orbital advancement (FOA) using LactoSorb® and found no instances of infection [9]. Similarly, Kurpad et al. found no infection in a review of 51 cases of LactoSorb® use in craniofacial surgery [10]. Sprecher et al. also found no instances of infection in 10 cases of laryngotracheal construction with LactoSorb® miniplates [11]. Lim et al. used LactoSorb® in 28 cases of elevation of a constructed auricle, and also reported no infection [12]. In another review by Tharanon et al. of 33 cases of LactoSorb® implantation in craniofacial surgery for craniosyostosis, hydrocephalus, cranial deformation and fibrous dysplasia, 1 instance of infection was found [13]. Kumar et al. found 1 instance of infection in 21 paediatric cases of LactoSorb® insertion for craniofacial surgery [14], and Goldstein et al. found no instances of infection among 8 cases [15]. However, both Kumar et al. and Goldstein et al.’s papers suffered from limited follow-up times (16 weeks and 3 months respectively), potentially underestimating the infection rate.
A total of 2685 patients had LactoSorb® implants in articles identified in this review, of whom only 18 patients developed infection (infection rate = 0.7%). Large prospectively designed studies included in this analysis corroborate these findings, indicating good evidence for low risk of infection in LactoSorb® use.
Biosorb® PDX (Bionx Implants Ltd, Tampere, Finland)
Biosorb® PDX is a blend of 80% PLLA and 20% PGA. One multicentre prospectively designed study by Ashammakhi et al. analysed outcomes in 161 paediatric patients with Biosorb® craniofacial implants and found 6 occurrences of post-operative infection (infection rate = 3.7%) [16].
PLLA (Generic)
Pure PLLA is a highly crystalline polymer, with an estimated resorption time of 3 years. Only two small articles were identified which analysed infection outcome in pure PLLA implants in paediatric surgery. Yagihara et al. found no instances of infection in 6 cases of PLLA meshes used for mandibular reconstruction [17]. Arai et al. found no instances of infection in 3 cases of PLLA implants used for craniosynostosis [18]. It should be noted that pure poly-l-lactic acid (PLLA) takes three years to be completely resorbed, and neither of these studies followed up all of their patients after three years had passed. Combined with their low sample size, this indicates that little can be inferred regarding PLLA’s infection rate in paediatric surgery.
One other article by Matsui et al. was identified; it analysed PLLA’s infection rate in a group of patients aged 5–19 years old, and found no cases of infection in 23 implants. However, the number of paediatric patients (0–18 years old) in that study could not be determined from the full text, so the infection rate of that study could not contribute to an analysis of PLLA’s infection rate in paediatric surgery [19].
MacroPore® PLA (Cytori Therapeutics, San Diego, CA, USA)
MacroPore® PLA is a blend of 70% PLLA and 30% poly-d,l-lactic acid (PDLLA), and has been used to hold the shape/position of weak bony structures, due to its high mechanical strength and its low impact toughness [20]. Cohen et al. (who worked for MacoPore® at the time of their study) reviewed 100 cases of MacroPore® PLA implantation for craniomaxillofacial surgery and found 4 instances of infection [21]. Branch et al.’s study of 203 cases of craniosynostosis repair included one group of 53 patients (inferred from percentage values in the text) who received MacroPore® PLA implants, 1 of whom developed infection [9]. A total of 153 patients had MacroPore® PLA implants in articles identified in this review, of whom 5 patients developed infection (infection rate = 3.3%).
PolyMax® (Depuy Synthes, Raynham, MA, USA)
PolyMax® is also composed of a blend of 70% PLLA and 30% PDLLA, but Imola et al. have suggested that different brands of PLLA/PDLLA (70/30) have different resorption profiles [22], so PolyMax® infection rates have not been analysed in combination with MacroPore® PLA infection rate. Bell et al. retrospectively reviewed 59 cases of facial fracture fixation with PolyMax® and found 1 case of infection (infection rate = 1.7%). The follow-up time ranged from 3 weeks to 3 years [23].
Inion® OTPS (Inion Oy, Tampere, Finland)
Inion® OTPS is a blend of trimethylene carbonate, PLLA and PDLLA. It has been used for surgical treatment of bone defects. Mavrogenis et al. retrospectively studied 9 cases of Inion® OTPS implantation for surgical treatment of congenital malformation, post-tumour excision reconstruction, fractures and osteotomies, and hand trauma with bone and soft tissue damage, and found no cases of infection after a minimum follow-up time of 7 months [24].
PolyMax® RAPID (Depuy Synthes, Raynham, MA, USA), RapidSorb® (Depuy Synthes, Raynham, MA, USA) and MacroPore® FRP (Cytori Therapeutics, San Diego, CA, USA)
PolyMax® RAPID, RapidSorb® and MacroPore® FRP are all made from a blend of 85% PLLA and 15% PGA. In a prospectively designed study by Cohen et al., 2 out of 168 patients who received a MacroPore® FRP implant for bone stabilization in craniofacial surgery developed an infection before the follow-up time (which ranged from 6 months to 3 years) [25]. In a prospectively designed review of 70 cases of frontal orbital advancement with RapidSorb® by Guzman et al., 1 patient developed an infection before the follow-up period (ranging from 3 months to 49 months) [26]. A retrospective review of 95 cases of RapidSorb® implantation for treatment of skull deformities carried out by Hayden Gephart et al. found 4 cases of infection (inferred from the percentage value found in the paper) [27]. In a review of 39 cases of Polymax® RAPID use for treatment of mandibular fractures, An et al. found no cases of infection [28]. In total, 372 patients received either PolyMax® RAPID, RapidSorb® or MacroPore® implants in articles identified in this review. 7 patients developed infection (infection rate = 1.9%).
Resorb-X® (KLS Martin, Tuttlingen, Germany)
Resorb-X® is commonly used during osteosynthesis, and is composed of 50% PLLA and 50% PDLLA. One study – by Freudlsperger et al.– found no cases of infection after a follow-up period between 15 and 21 months in a review of 38 cases of Resorb-X® implantation with the Sonic Weld system for craniofacial surgery [29].
Polydioxanone (Generic)
Polydioxanone is most commonly used as a biodegradable suture material. However, Polydioxanone bands can be used in orthopaedic and maxillofacial surgery. A single case series review – by Illi et al.– found no instances of infection in 32 cases of polydioxanone band implantation for craniofacial malformations, neuro-traumatic lesions and patellar osteo-chondral flake repair [30].
Tissue engineered vascular graft scaffolds
Synthetic materials used for vascular graft engineering are seeded with mononuclear cells before implantation into patients. The scaffolds can be composed of PLLA or PGA combined with 𝜀-caprolactone. One study by Hibino et al. found no cases of infection in 13 patients who underwent PGA vascular graft scaffold insertion, and no cases of infection in 9 patients who underwent PLLA vascular graft scaffold insertion. Both cohorts were followed up at 3 and 12 months, followed by a late-term follow-up at least 4.3 years after surgery [31].
Suprathel® (Polymedics, Denkendorf, Germany)
Suprathelr® is a resorbable synthetic skin substitute composed of a blend of polylactic acid, trimethylene carbonate and 𝜀-caprolactone. Highton et al. performed a prospectively designed study on use of Suprathel® in superficial partial thickness and mid-dermal burns in children, and found 4/33 patients developed infection (infection rate = 12.1%) [32]. This relatively high infection rate may be attributable to the high risk of infection in burns patients as opposed to the insertion of the biomaterial [33]. Madry et al.’s case series on the use of Suprathel® in a number of procedures, including partial thickness burns, did not report whether infection occurred after use on one 8-year old girl, but suggested infection rates significantly increase with delayed application of the Suprathel® patch [34].
Chi-squared test for independence
This article reviews the infection rates of 11 different materials. We performed a Chi-squared test for independence to determine whether or not infection rate varied across all materials. We reject the null hypothesis that infection does not vary with material across the studies included in this review (p < 0.0001). A significant result is robust to 1) considering only materials with non-zero infection rates (zero-infection rate materials had very low sample sizes); 2) considering all materials except Suprathel (this infection rate was a clear outlier). However, when considering all materials except LactoSorb® (used in more than half the sample), the result is no longer significant (p = 0.051).
To determine whether this represents a significant difference among LactoSorb® and all other materials (as opposed to only some materials), a t-test was carried out comparing LactoSorb®’s infection rate and the combined infection rate of all other materials. The result was statistically significant (p < 0.0001).
Individual comparisons between materials are beyond the scope of this review, due to problems with multiple hypothesis testing.
Discussion
Recently, there has been increased interest in the use of resorbable structures in surgery. The vast majority of surgeries identified in this study are related to repairing craniofacial defects. Traditionally, titanium implants were most commonly used for this type of surgery [29]. However, resorbable plates have become increasingly popular, largely due to the fact that the material disappears without the need for a second surgery which has its own risk of complications [35]. As well as craniofacial surgery, this review has identified use of synthetic resorbable materials in osteo-chondral flake repair, elevation of the auricle, vascular tissue engineering, laryngotracheal reconstruction, and skin repair.
Whilst the overall infection rate found in this review was low (41/3753 = 1.1%), evidence for each individual material’s use in paediatric surgery is highly variable. The material which has the strongest evidence to support its low risk of infection is LactoSorb®, demonstrated by large prospective reports with follow-up times encompassing total resorption of the material.
The Chi-squared test for independence showed statistically significant differences in infection rates among materials (p < 0.0001). There was no longer a significant difference when LactoSorb® was excluded from the comparison, and a t-test comparing LactoSorb® and all other materials suggested a lower infection rate for LactoSorb® than the other materials (p < 0.0001). This test should be interpreted with caution: the majority of the materials were not directly compared with each other in the primary research, and the quality of the studies being compared was variable. For example, type of procedure, follow-up times and study perspective differed between studies, all of which may confound the Chi-squared test. The implication of this result is that further research is needed to compare materials with each other in primary research controlling for the various confounders, which may well conclude more reliably that the material used influences the infection rate.
Infection is usually caused by perioperative contamination. However, it can occur any time after surgery before the material is resorbed. The risk factors for surgical-site infection include: co-morbidity (especially Diabetes Mellitus), increasing age, frailty, duration and complexity of surgery [36]. Many of these factors are often absent in the paediatric population, potentially contributing to the low infection rate found in this review. Specifically, with regards to biomaterials, bacteria first adhere to the material and then grow a biofilm on the surface of the material. Material surface charge and topography has been demonstrated to influence the likelihood of biofilm formation [37]. Further research into these characteristics of the different biomaterials presented in this review could help further explain the discrepancies in infection rate among materials. Resorbable plates are more frequently used in paediatric surgery than in adults. However, Noda et al. used LactoSorb for cranial fixation in adults and found no cases of infection, consistent with the low rates found in this review for paediatric cases [38].
Compared with Saxena et al.’s review of natural materials, this systematic review found a lower overall infection rate in synthetic materials (1.1% vs 5%). To address whether or not there is a real difference, original research needs to be carried out directly comparing synthetic and natural resorbable biomaterials in paediatric surgery, ideally prospectively designed with large sample sizes. Common to both reviews are a high infection rate in burns patients.
Materials implanted for tissue repair are designed to support the host’s tissue. Therefore, it is no surprise that other life, including bacteria, can exploit the presence of such materials for their own benefit. Gristina described this conflict between host tissue and pathogens as a “race for the surface” [39], and Busscher et al. called for more basic laboratory research to be done on the interaction among materials, host tissue and microorganisms to better understand how to minimize risk of infection for patients [3]. There is, therefore, a clear need for further study of infection in synthetic biomaterial implantation, from both laboratory and clinical research perspectives.
Conclusions
This review found a low risk of infection from paediatric surgery involving implantation of synthetic resorbable biomaterials. Particularly strong evidence was found for low infection rate in LactoSorb® use.
Footnotes
Conflict of interest
None to report.