Augmenting an Allograft for Anterior Cruciate Ligament Reconstruction With a Collagen Matrix and Bone Marrow Aspirate Concentrate Injection Appears Safe and Produces Favorable Clinical Outcomes at 2-Year Follow-Up

Purpose

To evaluate clinical outcomes of augmenting an anterior cruciate ligament (ACL) allograft with an amnion matrix wrap and injecting bone marrow aspirate concentrate (BMAC).

Methods

We enrolled 10 ACL reconstruction patients aged 22-60 years with hamstring allografts wrapped with an amnion collagen matrix and injected with BMAC in this prospective case series. Participants completed physical therapy and reported outcomes for 2 years. Postoperative magnetic resonance imaging scans were mapped/processed at 3, 6, 9, and 12 months, yielding mean transverse relations time constant (T2∗) and volume values for grafts and bone tunnel integration.

Results

We assessed the longitudinal outcomes of allograft augmentation using descriptive statistics and confidence intervals, showing significant increases in average Knee Injury and Osteoarthritis Outcome Score (KOOS ₅) and Single Assessment Numerical Evaluation (SANE) scores (KOOS ₅: Baseline-24 months [m] = 64.2-84.8, 95% confidence interval [CI] 9.14-32.56, SANE: Baseline-24m = 33.8-87.9, 95% CI 38.71-64.87). Average Veterans RAND 12 Item Health Survey (VR-12) Physical scores significantly increased from baseline to 24 months postoperation (Baseline-24m = 35.1-49.6, 95% CI 8.25-20.30). Average visual analog scale for pain scores significantly decreased from baseline at all time points except 2 weeks postoperation (baseline-2 weeks = 2.7-3.6, 95% CI −0.19 to 2.07) starting at 6 weeks postoperation (baseline-6 weeks = 2.7-1.3, 95% CI −2.52 to −0.26) and remained significantly lower than baseline to 24 months postoperation (baseline-24m = 2.7-0.6, 95% CI− 3.27 to −0.85). Average Max Activity Scale scores were significantly decreased from baseline starting at 12 months postoperation (baseline-12m = 6.3-2.7, 95% CI −6.32 to −0.88) but returned to baseline levels at 24-months postoperation (Baseline-24m = 6.3-5.8, 95% CI −4.45 to 1.45). There were no significant differences in VR-12 mental component scores or magnetic resonance imaging measures. No infections nor reconstruction failures occurred after 2 years.

Conclusions

This case series demonstrated augmenting hamstring allograft ACL reconstruction with an amnion collagen matrix and injecting BMAC appeared to be safe, and clinical outcomes were favorable up to 2 years postoperation despite having no quantifiable effect on graft maturation.

Level of Evidence

Level IV, therapeutic case series.

The use of allograft tissue for reconstruction of the anterior cruciate ligament (ACL) remains popular primarily as the result of decreased postoperative pain and morbidity, shortened surgical times, and improved cosmesis. ^,^, However, the increased failure rate seen in younger and more active patients has kept its use limited in favor of autograft options. The graft failure cause is often multifactorial. However, additional etiologies of failed ACL reconstruction include malposition of the femoral tunnel, recurrent injury of the knee, or premature return to activity. This increased failure rate has been attributed to the longer graft maturation rates in allograft tissue grafts. ^, Although ACL allografts heal through the same biologic steps as autografts, they do so at a slower rate and lose more of their initial strength than autografts in the healing process. ^,

The application of biologics to enhance graft healing, including platelet-rich plasma, bone marrow aspirate concentrate (BMAC), and cellular therapies has been examined in conjunction with ACL reconstruction, and there has been no consensus on neither the efficacy nor the best method of application. ^,^,^,^,^,^, The question remains: is it the type and preparation method of the biologic adjunct, or is it the technique of application, or a combination of both? The principles of augmentation with biologics have classically centered on the triad of tissue, cells, and scaffolds. ^,^, Because the normal uninjured ACL has a synovial layer, which contributes to the blood supply and nutrition of the native ACL, and possibly protects it from the unique intra-articular environment, the addition of a protective scaffold in ACL repair and reconstruction has been investigated and appears promising. ^,^,^, The use of such a membrane or scaffold is also a method to contain the added cells and growth factors, such as hepatocyte growth factor, epidermal growth factor, tumor necrosis factor-α, tissue inhibitor of metalloproteinases, insulin-β like growth factor, interleukin 1-receptor antagonist, and transforming growth factor (α, β1, β2), near the target tissue. Amniotic tissue-derived membranes have been examined in ophthalmology, oral surgery, and other wound-healing applications. ^,^,^,^,^, The role of these membranes as a contributor to an intra-articular healing process is unknown, but they have shown promise as a barrier and scaffold. ^,^,^,

The recent scientific progress in understanding the principles and potential applications of orthobiologics, along with increased consumer demand, has naturally led to the question of whether biologic augmentation of ACL grafts can increase the maturation rates of allograft ACL tissue, or whether augmented allografts can improve maturation at the graft-tunnel interface. There are few studies that have biologically augmented ACL allografts. ^,^,

The purpose of this study was to evaluate the clinical outcomes of augmenting an ACL allograft with an amnion matrix wrap and injecting BMAC. We hypothesized that this method of augmenting ACL reconstruction surgery would produce favorable clinical outcomes, as shown by the primary effectiveness end point, patient-reported outcome measures (PROMs) including Knee Injury and Osteoarthritis Outcome Score (KOOS), visual analog scale (VAS), Single Assessment Numerical Evaluation (SANE), Marx Activity Scale (MAS), and Veterans RAND 12-Item Health Survey (VR-12) physical and mental components.

Methods

Participants

This was a prospective case series with 2 years of follow-up. Institutional review board approval was obtained (number 1309783, Baptist Hospital Institutional Review Board). Enrollment was started September 2018, completed in July 2019, and was executed in the outpatient setting at the primary institute (Andrews Research & Education Foundation). Follow-up completed in July 2021, and final data analysis was completed by September 2023. No changes were made to neither the methods nor outcome measures after trial commencement.

Patients between the ages of 22 and 60 years who were scheduled to have primary ACL reconstruction with QuadLink Allograft (Arthrex, Naples, FL) by one of the investigating physicians were screened for participation in the study by members of the research team. A total of 10 participants were included in the case series, 6 male (60%) and 4 female (40%) patients. The age range was 37-47 years with a mean ± standard deviation of 43.30 ± 3.56 years. Exclusion criteria included patients with previous procedures such as revision surgeries or significant previous injuries to the same knee, patients with difficulty obtaining internet access, patients who did not have an active e-mail address, and patients who could not comprehend study documents or give informed consent. Incidental concomitant injuries were not exclusionary to the study to capture a participant group representative of a healthy, active population. Patients who were unable to complete magnetic resonance imaging (MRI) examinations because of claustrophobia or anxiety also were excluded. Potential participants were informed in a neutral manner of all potential benefits and complications of participating in the study. If eligible and interested in participating, the potential participants went through the informed consent process (enrollment) with a member of the research team. No specific advertising or recruitment materials were used. No compensation was given to participants in this study. Subjects also were enrolled in the Surgical Outcomes System (SOS; Arthrex) knee arthroscopy registry and completed all preoperative questions before the surgical date.

Intraoperative Procedures

The surgical procedures were independently performed by 2 surgeons (A.W.A. and S.E.J.). All participants underwent an ACL reconstruction with a QuadLink Allograft (Arthrex). A similar method of reconstruction was used in all cases, using minimal notchplasty and an all-inside technique using adjustable suspensory cortical fixation augmented with an internal brace construct, Arthrex TightRope (Arthrex). Before starting the arthroscopic procedure, bone marrow aspirate was harvested from the posterior iliac crest in each case. The bone marrow harvest was processed using the Arthrex Angel blood processing system using the 15% hematocrit setting. One milliliter of BMAC was removed and sent for analysis, including colony-forming unit (CFU) and total nucleated cells (TNC) counts, immediately after harvest. The remaining BMAC was transferred onto the field in a sterile fashion to preserve sterility of the BMAC.

During graft preparation, a FiberTape was attached to the femoral suspensory button to be used as suture augmentation. After each graft construct was readied and placed under tension on a graft prep stand, they were placed in compression tubes. Just before the time for implantation of the graft, each graft was removed from the compression tube and was wrapped with a 3-cm by 6-cm amnion collagen matrix (Arthrex Amnion Matrix Thick; Arthrex). The length of the matrix allowed for 3 wraps, creating a cylinder of matrix around the graft. Due to tightly wrapping the thin collagen matrix (less than 0.5 cm), this procedure added an inconsequential amount to the thickness of the graft. Using a 4-0, poliglecaprone-25, monofilament suture (MONOCRYL; Ethicon, Raritan, NJ), a cerclage suture was placed at each horizontal end of the cylinder and a running non-locking suture was run along the vertical end of the wrap ( Fig 1 A ). This created a watertight barrier around a 3-cm section of the intra-articular portion of the grafts.

Image, AltText currently not available — Fig 1

Before graft implantation, each tunnel socket, approximately 9 to 10 mm in diameter, was filled with demineralized bone matrix to allow graft tunnel interface integration (AlloSync Pure; Arthrex). After graft implantation, which included tensioning of the graft independent of securing the fiber tape distally, the joint was dried and the remaining BMAC (2 mL on average) was injected into the watertight cylinder under the membrane with a 22-gauge spinal needle under direct arthroscopic visualization ( Fig 1 B and C). Postoperative rehabilitation protocols followed the institution’s standard for ACL reconstruction with variation as needed dependent on associated pathology or meniscus work.

Postoperative Follow-Up

All subjects in the case series were followed with an online PROMs system. The primary outcome measure was a change in the PROMs from preoperative to postoperative time points. The outcome system used e-mail prompts and online questionnaires to track patient-reported outcomes. PROMs were captured with 5 questionnaires, the SANE, MAS, KOOS, the VAS Pain, and the VR-12 physical and mental means on the schedule presented in Figure 2 . ^, In addition, participants reported adverse events such as swelling, stiffness, graft rupture, or allergic reaction to assess the safety of the intervention.

All subjects also underwent postoperative MRI at the 3-, 6-, 9-, and 12-month postoperation time points, which Arthrex funded. Each MRI used T2∗ sequences and data processing was performed to acquire a mean T2∗ value and volume for each graft at every time point. The steps followed by the investigator in the processing have previously been described and validated to establish intra-rater reliability. These values have been shown to detect differences in ACL content, structure and maturation. ^, Mapping involved manually drawing a region of interest around the ACL on each image where the ACL was visible on the 3D true FISP (fast imaging with steady-state free procession) series, which was performed by a sports medicine−trained orthopaedic surgeon with MRI mapping experience (A.W.A.). ^,^, This is represented in Figure 3 . The true FISP series was selected because previous authors have asserted that this sequence best highlights the difference between the ACL and surrounding structures, including synovial fluid. Segmenting was performed in OsiriX (Pixmeo, Geneva, Switzerland) ( Fig 3 ).

The MRIs were deidentified before being analyzed. A 3-parameter single exponential decay model was fit to the T2∗ data in each voxel of the ACL segmentation.

y = e ( − t B ) + C

Volume was calculated by summing the total number of ACL graft voxels. The segmentations defined on the 3D True FISP could not be directly used on the T2∗ mapping images because they had differences in slice prescriptions and voxel sizes. Consequently, custom software written in MATLAB (MathWorks, Inc., Natick, MA) was used to account for these differences while transferring the segmentations to the T2∗ mapping images. Customized software using MATLAB (MathWorks) was used to calculate the mean T2∗ values and standard deviation for each clinically relevant subregion.

The graft tunnel interface integration was visually analyzed in the method of Gupta et al., using a 5-point scale by a sports medicine−trained orthopedic surgeon (S.E.J.). Interface signal intensities were scored 5 when similar to joint fluid, 4 when greater than muscle but less than joint fluid, 3 when similar to muscle, 2 brighter than tendon but less than muscle, and 1 if the signal was similar to patellar or quadriceps tendon. Gupta Sagittal STIR images were used and the predominant signal intensity (>50% of the graft surface) between the graft and the bone tunnel was recorded at each time point, 3, 6, 9 and 12 months.

Statistical Analysis

A sample size of 10 participants was chosen to complete this study, which is generally the maximum sample size for a case series. All demographic data and MRI integration data were entered into Excel (Microsoft, Redmond, WA). MRI mapping was completed in OsiriX (Pixmeo) and exported to Excel. Descriptive statistics (mean and standard deviation) were reported for all numerical outcomes (MRI and PROMs data). For the patient reported outcomes (PROMs), 95% confidence intervals (CIs) were calculated and presented using RStudio (2021). ^,^,

Results

Analysis of the bone marrow aspirate product demonstrated 139.7 ± 20.9 (10.3-311.0) TNC (k/μL) and 410.9 ± 143.8 (0-2,340) CFU/mL for the 10 participants of this study. Bone marrow concentrate results reflected normal healthy population results (TNC: 39.7 ± 17.2 million/mL to 66.1 ± 20.5 million/mL, CFU-fibroblasts/mL 500-3,000). All participants received the allocated treatment and completed the baseline questionnaires. There were no postoperative complications (infections, stiffness, persistent effusions), BMAC morbidity, or reconstruction failures, graft ruptures confirmed through imaging, after 2 years of follow-up for any of the participants. No participants required concomitant procedures or adjustments to the rehabilitation protocol.

The average KOOS _5, SANE, and VR-12 physical component scores were significantly increased from baseline at all time points and remained significantly greater than baseline out to 24 months postoperation. The PROMs data are included in Tables 1-2 and Figure 3 . Importantly, the KOOS ₅ and SANE improvements were observed at 3 months postoperation, which is well before the standard return to activity time at 6 months postoperation. The improvement in average VR-12 physical component scores was observed at the 6-month time point. All functional improvements were maintained out to 2 years postoperation whereas the physical activity of participants decreased based on the average MAS scores. It is worth noting that baseline physical activity levels did not return at 12 months postoperation and were approaching baseline at 24 months of follow-up. Pain levels, represented by the average VAS Pain scores, were significantly decreased at 6 weeks postoperation. The average VAS Pain scores and MAS scores were significantly decreased from baseline at all time points except 2 weeks postoperation and the VAS remained significantly lower than baseline out to 24 months postoperation, whereas the MAS returned to baseline levels at 24 months postoperation. Improvement in pain levels was maintained out to 24 months postoperation along with increases in average physical activity levels. The average VR-12 mental component scores cores were not significantly different from baseline at any time point, including 24-months postoperation (baseline to 24 months = 56.5-57.4, 95% CI −5.02 to 8.3, P =.611). Figure 4 presents PROMs outcomes scores across time points.

Table 1

Patient-Reported Outcome Measures (PROMs)

	Baseline	2 Weeks	6 Weeks	3 Months	6 Months	1 Year	2 Years
SANE, mean (95% CI), (0-100)	33.8 (16.5- 51.1) n = 10	−	−	65.9 (52.4- 79.4) n = 10	74.1 (67.0-81.2) n = 10	80.2 (68.9- 91.5) n = 9	87.9 (81.9-93.9) n = 8
MAS, mean (95% CI)	6.3 (18.7- 50.7) n = 10	−	−	−	−	2.7 (82.2- 91.8) n = 10	5.8 (81.5- 95.5) n = 8
KOOS ₅, mean (95% CI), (0-100)	64.2 (52.2- 76.3) n = 10	−	−	77.1 (67.3- 87.0) n = 10	75.6 (65.2- 86.0) n = 10	80.6 (66.6- 94.5) n = 9	84.8 (71.0- 98.6) n = 8
VAS, mean (95% CI), (0-10)	2.7 (0.9-4.5) n = 10	3.6 (2.3- 5.0) n = 10	1.3 (0.5-2.2) n = 10	0.9 (0.3-1.5) n = 10	0.7 (0.1-1.4) n = 10	1.1 (0.3-1.9) n = 10	0.6 (0.0-1.3) n = 8
VR-12 physical, mean (95% CI)	35.1 (27.4- 42.8) n = 10	−	−	−	45.2 (39.3-51.1) n = 10	47.1 (41.8- 52.4) n = 9	49.6 (43.3- 55.9) n = 8
VR-12 mental mean (95% CI)	56.5 (49.8- 63.2) n = 10	−	−	−	53.1 (46.3- 59.9) n = 10	55.8 (49.6- 62.0) n = 9	57.4 (53.4- 61.4) n = 8

CI, confidence interval; KOOS, Knee Injury and Osteoarthritis Outcome Score; MAS, Marx Activity Scale; SANE, Single Assessment Numeric Evaluation Score; VAS, visual analog scale for pain; VR12, Veterans RAND 12 Item Health Survey.

Table 2

Linear Mixed-Effects Models for PROMs Survey Observation Data (Reference Time Is Baseline)

Predictors	KOOS			SANE
Predictors	Estimates	95% CI	P Value	Estimates	95% CI	P Value
(Intercept)	64.24	54.08-74.41	<.001	33.80	23.13-44.47	<.001
Time [3M]	12.88	1.96-23.79	.022	32.10	19.90-44.30	<.001
Time [6M]	11.36	0.44-22.28	.042	40.30	28.10-52.50	<.001
Time [12M]	16.36	5.08-27.64	.006	45.68	33.07-58.28	<.001
Time [24M]	20.85	9.14-32.56	.001	51.79	38.71-64.87	<.001
n _Patients = 10, n _{survey observations} = 47
Predictors	VAS			MAS
Predictors	Estimates	95% CI	P Value	Estimates	95% CI	P Value
(Intercept)	2.70	1.76-3.64	<.001	6.30	2.58-10.02	.002
Time [2W]	0.94	−0.19 to 2.07	.101
Time [6W]	−1.39	−2.52 to-0.26	.017
Time [3M]	−1.78	−2.91 to −0.65	.003
Time [6M]	−1.96	−3.09 to −0.83	.001
Time [12M]	−1.61	−2.74 to −0.48	.006	−3.60	−6.32 to −0.88	.013
Time [24M]	−2.06	−3.27 to −0.85	.001	−1.50	−4.45 to 1.45	.297
n _Patients = 10, n _{survey observations} = 68				n _Patients = 10, n _{survey observations} = 28
VR-12 Physical				VR-12 Mental
Predictors	Estimates	95% CI	P Value	Estimates	95% CI	P Value
(Intercept)	35.10	29.57-40.64	<.001	56.48	51.03-61.93	<.001
Time [6M]	10.09	4.49-15.68	.001	−3.39	−9.61 to 2.83	.271
Time [12M]	11.79	5.99-17.58	<.001	−0.03	−6.47 to 6.40	.991
Time [24M]	14.28	8.25-20.30	<.001	1.67	−5.02 to 8.3	.611
n _Patients = 10, n _{survey observations} = 37