Retrospective Cohort Analysis on the Use of Hydrolyzed Collagen Powder in Back Reconstruction Following Spinal Instrumentation

© 2024 HMP Global. All Rights Reserved.

Any views and opinions expressed are those of the author(s) and/or participants and do not necessarily reflect the views, policy, or position of ePlasty or HMP Global, their employees, and affiliates.

Posterior spinal fusion is a common surgical procedure employed for the treatment of a wide range of pathologies, ranging from congenital to trauma to degenerative changes and malignancy. Medical and surgical risks of the procedure are high and can be predicted on the basis of comorbidities and operative factors.^1,2 Infections are estimated to occur in 0.7% to 20% of cases, and much-feared hardware failure happens as often as 1.9% to 12.5% of the time in high-risk cohorts.^3,4 The 90-day readmission rate following lumbar fusion is 9.7% in a large national survey.⁵ In cases of revision spinal fusion procedures, there is an increased risk of surgical site infections and dural tears compared with that of primary procedures.⁶

Due to the relative frequency of wound-related complications in spinal procedures, plastic surgeons are commonly employed to reconstruct defects and increasingly are consulted at the time of the index operations. Plastic surgery involvement in the closures of these procedures is related to a statistically significant decrease in the incidence of postoperative infections and 30-day readmission rates.⁷ Complex reconstruction with local muscle flaps is being used by plastic surgeons to lower complication rates in very high-risk groups.^8-10 Similarly, in a systematic review, prophylactic use of well-vascularized flaps for coverage of spinal wounds is associated with a decreased number of wound complications in those with instrumentation.¹¹

In many cases where multiple levels are involved, the surrounding soft tissues can be undermined widely to achieve a satisfactory closure but at the expense of creating vast dead spaces that fill with fluid in the postoperative period. This may explain a higher rate of seroma formation for spinal closures performed by plastic surgeons than is seen with those performed by spine surgeons in 1 comparative analysis (seroma rate of 36.5% vs 3.8%, respectively).¹² It is worth noting that in an earlier cited study on the benefits of plastic surgery spinal closures, the rate of postoperative seroma was recorded as only 0.5%.⁷ The ability of plastic surgeons to avoid these adverse events is paramount to delivering consistent results and optimizing patient outcomes.

In the areas of breast surgery and abdominal wall reconstruction, both of which are plagued by frequent seromas, there have been promising results in the use of collagen or hemostatic powders to prevent their formation.^13-17 To this end, we sought to study the success of a collagen-based powder in the reduction of postoperative seromas following reconstruction of instrumented spinal wounds.

Study Design

A retrospective review of cases from the same surgical team consisting of neurosurgery and plastic surgery attending physicians was performed for patients undergoing spinal fusion with local muscle flaps over a span of 20 months. Demographic factors, including age and sex, as well as pertinent medical and surgical history were recorded. Intraoperative considerations, including the use of hardware and number of involved spinal levels, were noted. Cohorts were grouped on the basis of the use of CellerateRX Surgical hydrolyzed collagen powder (HCP; Sanara MedTech) within the wound bed. The use of this powder represented a practice shift for the principal investigator (D.K.) whereby those patients for whom the compound was used could be compared with the initial cohort for whom no such compound was used. The protocol was reviewed by the institutional review board of our institution, and exemption was granted for this retrospective review.

Surgery Overview

Spinal fusion with instrumentation proceeded based on the standard practice of the neurosurgical team. Following the conclusion of their procedure, paraspinal muscle flaps were raised by the plastic surgery team, taking care to free the muscle posteriorly and anteriorly to achieve a midline tension-free closure. In cases of the latter cohort, the hydrolyzed collagen powder was placed within the wound bed. A round silicone drain was placed deep to the paraspinal muscle and fascial closure. The muscle was then approximated in the midline, providing coverage over the spinal hardware. A superficial drain was then placed in a similar fashion above the muscle. A layered skin closure was performed, and the incision was dressed with a semi-occlusive sterile dressing.

Outcomes

Primary outcomes assessed within the review included the rates of seroma, infection, wound dehiscence, and return to the operating room. Secondary outcome measures included drain outputs, the length of time until final drain removal, and length of hospital stay postoperatively. These measures were used to compare the above cohorts. Follow-up through analysis of inpatient and outpatient records was carried out for a period of 3 months postoperatively.

Data Analysis

Quantitative statistical analysis was used to draw conclusions and inform comparisons between patient groupings. For comparisons between 2 cohorts, a Student t test was used to compare means. For categorical comparisons, chi-square analysis was performed. For the purpose of the aforementioned statistical tests, associated P values below .05 were deemed significant. All statistical calculations were performed using Excel v2107 (Microsoft Corp).

A total of 76 patients were included within the study period after undergoing a spinal fusion procedure along with reconstruction by plastic surgery with paraspinal muscle flaps. Of these, 29 patients were within an initial cohort treated with standard of care closure with drains, and 47 patients were subsequently included within an experimental cohort that included the use of HCP in addition to the standard of care treatment. Table 1 compares the cohorts on the basis of demographic and surgical factors. On average, patients in the control group tended to be a few years younger than their counterparts in the experimental cohort, and this achieved statistical significance (P = .04). With regards to sex, mean body mass index (BMI), history of hypertension, history of diabetes, current smoking status, history of prior spinal procedure, or the number of involved spinal levels, there were no statistically significant differences between the cohorts.

There were a greater number of clinically apparent seromas requiring drainage in the experimental group than in the control group (13 versus 2, P = .03) but no clinically significant hematomas requiring drainage or return to the operating room in either group. There was a tendency towards a higher rate of wound dehiscence, infection, and return to the operating room for the experimental group, although none of these differences were statistically significant. Table 2 provides an overview of these comparisons. Groups did not differ significantly also with respect to the average postoperative length of stay or the number of days until final drain removal. Figure 1 provides a graphic comparison of the difference in average daily drain outputs between the cohorts for the first week postoperatively. These differences did not achieve statistical significance.

Figure 1. A comparison of average daily drain outputs between cohorts. For the first postoperative day, the period reflects the amount of time from the end of surgery until the next morning at the change of the first nursing shift.

Further analysis was performed to examine the relationship between the recorded demographic and surgical factors and the occurrence of seromas, regardless of grouped cohort. Figures 2 and 3 highlight case examples of the scope of the defects and the resulting morbidity from a seroma, respectively. In the present study, there was a tendency towards a higher rate of seromas with such factors as increased age, current smoking, reoperation, and history of diabetes, though none of these differences reached statistical significance (P > .05). Likewise, there was an association between seromas and higher average daily drain outputs, but this was not found to be statistically significant (P = .19). Table 3 summarizes the above comparisons in detail.

Figure 2. Intraoperative photograph from a case of a patient undergoing revision of lumbar spinal fusion with removal and replacement of hardware for treatment of persistent back pain. The scope of the wide defect can be appreciated along with the large amount of potential space available for fluid to collect even with minimal undermining and closure.

Figure 3. Postoperative photographs for a 20-year-old patient who underwent 2-stage correction of severe kyphosis with hardware placement. On the left, the photograph shows the incision line after removal of the dressing and both surgical drains on postoperative day 4 following the second stage. The patient was discharged that day with no collection noted. The patient returned to the emergency room on postoperative day 8 with the superficial fluid collection appreciable at the lower aspect of the incision in the photograph on the right.

With a continually high volume of spinal procedures performed across the country on a yearly basis, there are many opportunities for plastic surgeons to collaborate to optimize outcomes for patients. In many cases, attention to principles of soft tissue handling and vascularized coverage can be paramount to prevent dreaded hardware exposure and life-threatening infections. The wide undermining that is often employed for a tension-free closure puts the patient at risk of postoperative seromas that can lead to wound breakdown and infection. The ability to mitigate these risks is fundamental. It has long been recognized that the harvest of muscles from the back, such as the latissimus dorsi, leads to high rates of postoperative seromas, and many attempts have been made to prevent these complications through the use of sealants and quilting sutures, among other modalities.^18-20

In addition to fibrin and thrombin-based products, collagen-based products have been used for a similar purpose in accelerating healing. Collagen-based particulates are used to facilitate complex wound healing through providing the building blocks for extracellular structures, hemostasis, and recruitment of cells involved in the wound healing cascade.^21-22 The proposed mechanism of action for the use of collagen in wound healing is twofold. Through signal interactions, it helps to draw in cells, such as fibroblasts, to allow for structured healing and the creation of new extracellular matrix.²¹ In addition, the added collagen “martyrs” itself by undergoing breakdown by matrix metalloproteinases (MMPs) and thereby preserving the native extracellular matrix.²¹ The latter point is more applicable in chronic wounds where circulating MMPs interrupt the natural attempts to rebuild extracellular frameworks. On a macroscopic scale, collagen-treated wounds exhibit organized granulation tissue when compared with untreated wounds, also confirming its role in cell recruitment and organization for proper wound healing.²²

The hypothesis for the present study was based on this premise that the addition of collagen to the surgical wound bed would facilitate more rapid healing between tissue structures through the accelerated repair of extracellular attachments, thereby avoiding the formation of large fluid-filled pockets. In the present study, this hypothesis was challenged, and there was a significantly greater rate of seromas in the experimental arm treated with the collagen powder. There was a trend towards a higher rate of wound dehiscence and infection in the experimental arm as well, though these differences were not clinically significant. Dehiscence and infection can result from untreated seromas, though the seromas detected in the study were typically drained promptly when identified to prevent further complication.

There are few studies of note in the published literature on the use of collagen additives for wound healing. Qureshi et al reported on the use of collagen applied topically to a punch biopsy wound allowed to heal secondarily, and the authors compared outcomes with those for the same wound type closed primarily. In the study, the majority of experimental wounds treated with collagen (75.0%) healed within 4 weeks, had improved cosmesis, and had histologic evidence of more organized granulation.²² Kumar et al also reported on the use of HCP applied topically for wounds and showed that treatment promoted faster re-epithelialization and improved breaking strength of healed scars.²³ The group noted that HCP promoted expression of anti-inflammatory interleukin-10 as well as proangiogenic vascular endothelial growth factor.²³ Stafyla et al randomly assigned patients undergoing axillary node dissection to either receive collagen powder in the wound bed or to proceed in standard fashion with no sealant. The group found that patients treated with the collagen powder tended to have decreased drainage, but results were not significant.²⁴ Finally, Akopov et al reported on the treatment of malignant pleural effusions with drainage and the use of collagen powder to achieve lasting pleurodesis in 89.0% of patients, but there was no control group in this study.²⁵ Taken together, HCP appears to induce structured healing and re-epithelialization when used externally. When used internally to reduce dead space, there is a lack of evidence to support its use, and further study is needed.

There are important limitations to note for this study. The present study appeared to be underpowered in many of the analyses, with few comparisons showing statistically significant differences. There were a number of trends in the present study, but these could not be supported on the basis of the evidence presented. Additionally, the experimental cohort was on average slightly older than the control cohort, and advanced age is associated with seroma formation in the literature.²⁰ That said, when performing a post hoc analysis analyzing differences in patients who did or did not develop seromas, none of the demographic or surgical factors differed by a statistically significant degree between the latter groups.

Ultimately, large high-quality studies are needed to clarify the role of HCP and other sealants. On the basis of this study, it would not be possible nor prudent to recommend the continued use of HCP in these surgeries without further analysis. Further research can also be done to explore a possible role of collagen powder in seroma formation. It is possible that in the way the collagen particulates facilitate healing in external wounds, they lead to a similar barrier formation internally between isolated tissue planes to counterproductively seal off a fluid cavity.

In light of the ambiguous findings, our group has discontinued the use of collagen powder within the context of this procedure. Tested recommendations in the literature for preventing seroma formation can be used in place, and ultimately prophylactic measures aimed at minimizing undermining may be most helpful. In conclusion, this study presents evidence that the use of a collagen-based powder in complex back reconstruction does not adequately prevent postoperative seromas and may in some cases lead to a greater number of them. Larger, multicenter studies performed prospectively with randomization are ultimately needed to draw definitive conclusions. In the meantime, the most prudent measures may be ones aimed at limiting the scope of the problem from the outset.