Significant improvements in clinical outcome measures and patient satisfaction after combined all-arthroscopic meniscal allograft transplantation and autologous chondrocyte implantation: A single-centre longitudinal study

Abstract

Purpose

The optimal treatment approach for the complex pathology of meniscal insufficiency and coexisting full-thickness cartilage defects remains unclear. The purpose of this study was to evaluate the viability, safety, and efficacy of this combined surgical approach at medium-term follow-up.

Methods

This is a single-centre longitudinal study with blinded outcome assessment. All consecutive patients treated with combined all-arthroscopic meniscal allograft transplantation (MAT) with bone bridge fixation and ACI using chondrospheres at our institution between 2001 and 2021 were eligible for inclusion. Twenty patients with an average follow-up of 72.6 ± 34.4 months were included in the statistical analysis. Clinical outcomes were assessed pre- and postoperatively using the IKDC Subjective Knee Form, Lysholm Score, Tegner Activity Scale, KOOS, and Visual Analog Scale (VAS) for patient satisfaction. Failure and reoperation rates were assessed, and cartilage regeneration tissue was evaluated on postoperative MRI.

Results

IKDC scores significantly improved from 52.1 ± 16.9 to 68.5 ± 16.3 (p = 0.003). Lysholm scores improved from 61.5 ± 21.7 to 78.5 ± 12.9 (p = 0.004). Tegner scores improved from 3.5 (1–4) to 4.0 (2–6) (p = 0.014). KOOS scores improved significantly across all subcategories, except ‘symptoms’, where improvements did not reach statistical significance. VAS for overall patient satisfaction showed improvements but did not reach statistical significance. The combined procedure was successful in 17 patients (85%). Eight patients had to undergo reoperation (40%), comprising mostly small, arthroscopic procedures. Seven reoperations were directly attributable to meniscal allograft transplantation (46.7%). Postoperative Magnetic Resonance Observation of Cartilage Repair Tissue scores were 68.9 ± 16.8 (n = 14).

Conclusion

Combined arthroscopic MAT and autologous chondrocyte implantation (ACI) is a viable, safe, and effective treatment approach for younger patients with meniscal insufficiency and coexisting full-thickness cartilage damage, where alternative treatment options are limited. The combined surgical procedure achieved significant improvements in clinical outcome measures and patient satisfaction with acceptable failure and high arthroscopic reoperation rates. MAT is the limiting part of this combined procedure, with most failures and reoperations being attributable to MAT, as opposed to ACI.

Level of Evidence: Level III.

Abbreviations

ACI

autologous chondrocyte implantation

ADL

activities of daily living

AMIC

autologous matrix-induced chondrogenesis

BMI

body mass index

IKDC

International Knee Documentation Committee

KOOS

Knee Injury and Osteoarthritis Outcome Score

MAT

meniscal allograft transplantation

MFC

medial femoral condyle

MFx

microfracture

MOCART

magnetic resonance observation of cartilage repair tissue

MRI

magnetic resonance imaging

PRISMA

preferred reporting items for systematic reviews and meta-analyses

QoL

quality of life

SD

standard deviation

STROBE

Strengthening the Reporting of Observational studies in Epidemiology

VAS

visual analogue scale

INTRODUCTION

Meniscal injuries are a common diagnosis, with an estimated prevalence of 222 per 100,000 [1], often going hand-in-hand with cartilage lesions [2]. Despite advances in surgical meniscal restoration, numerous lesions remain irreparable [3], often necessitating meniscectomy [4-6], which can accelerate joint degeneration, leading to the development of osteoarthritis [7]. Meniscal allograft transplantation (MAT) and autologous chondrocyte implantation (ACI) are surgical interventions developed to address these conditions without resorting to arthroplasty. Extensive evidence has demonstrated these procedures to be safe and effective when performed independently [8-17]. However, conclusive evidence on combined MAT and ACI is still pending, and only few studies have addressed this topic [18-22].

Historically, full-thickness cartilage defects were considered contraindications for MAT [23], but recent evidence has started to challenge this notion, advocating for a more comprehensive approach to address all existing pathologies, including multipronged surgical approaches such as combined MAT and ACI [11, 21, 24-26]. A 2021 systematic review [27] highlighted conflicting evidence concerning the impact of chondral damage on MAT outcomes. However, a more recent extensive systematic review and meta-analysis [28] identified a pronounced association between the extent of cartilage damage at the time of MAT and subsequent graft failure. Despite these findings, only a few studies have reported outcomes after combined articular cartilage restoration and MAT, demonstrating the potential for positive results and warranting further investigation [29]. However, despite growing evidence of the potential benefits, few studies have investigated combined MAT and ACI [18-22], with considerable heterogeneity in methodology and outcome reporting [29].

This limited and heterogeneous research landscape indicates a clear need for further investigation. Thus, we provide a detailed exploration of this combined arthroscopic approach, contributing to a deeper understanding of its potential benefits and limitations. The findings could guide clinical practice and inform future clinical decision-making when faced with combined meniscal and chondral pathologies.

A longitudinal study was designed with the aim of evaluating the viability, safety, and efficacy of combined MAT with bone bridge fixation and ACI using chondrospheres in patients with meniscal insufficiency and coexisting full-thickness cartilage defects. The primary objective was to assess the clinical outcomes after combined MAT and ACI at medium-term follow-up. Secondary objectives were to evaluate patient satisfaction, failure rates, reoperation rates, and postoperative cartilage regeneration tissue. The hypothesis was that combined MAT and ACI would result in significant, clinically meaningful improvements in clinical outcome measures for younger patients with coexisting loss of meniscal function and cartilage damage, where biologic treatment options are limited.

MATERIAL AND METHODS

Study design and regulatory information

After obtaining institutional review board approval (ATOS Clinic Heidelberg, ethics committee, reference number: 12021), a single-centre longitudinal study with blinded outcome assessment was conducted at the International Center for Orthopedics, ATOS Clinic Heidelberg, a tertiary referral institution for orthopaedic surgery and sports traumatology. The study aimed to (i) evaluate the viability, safety, and efficacy of combined MAT and ACI and (ii) assess failure and reoperation rates of this combined procedure.

Written informed consent was obtained from all patients. All patients were over the age of 18 at the time of inquiry (the age range at the time of index surgery was 15–47 years). This study adhered to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines for reporting observational research [30] and the Declaration of Helsinki [31]. Data supporting our findings are available from the corresponding author upon reasonable request.

Patient selection, eligibility criteria, and outcome assessment

All consecutive patients treated with combined all-arthroscopic MAT with bone bridge and ACI using chondrospheres at our institution between 2009 and 2021 by the senior surgeon (R.S.) were included. Table 1 details inclusion and exclusion criteria, and Figure 1 summarises patient flow during screening and selection.

Table 1. Study inclusion and exclusion criteria.

Inclusion criteria	Exclusion criteria
Total or subtotal loss of medial or lateral meniscus and concomitant chondromalacia (Outerbridge 3–4°) affecting the ipsicompartmental femoral condyle and/or tibial plateau	Relevant complex trauma to the affected knee joint
Surgical treatment with combined MAT and ACI	Rheumatoid comorbidity
Age at index surgery 14–50 years	Neurological diseases
Signed written informed consent by patient or legal guardian	Follow-up duration of less than 12 months
Normal lower extremity alignment	Relevant ligamentous instability

• Abbreviations: ACI, autologous chondrocyte implantation; MAT, meniscal allograft transplantation.

Figure 1

Preferred reporting items for systematic reviews and meta-analyses (PRISMA) flow chart. Standard PRISMA flow chart detailing patient flow during the screening and selection process.

Two independent raters (I.F., S.F.), not involved in the surgical procedure and blinded to patient identity and type of surgery performed, assessed patients using internationally standardised and validated questionnaires both pre- and postoperatively. Data were collected according to a predetermined protocol. The following patient-reported outcome measures were collected: IKDC [32], Knee Injury and Osteoarthritis Outcome Score (KOOS) [33], Tegner Activity Scale [34], Lysholm Score [34], and Visual Analogue Scale (VAS) [35] for patient satisfaction. Failure and reoperation rates were assessed as well. Chondral defect size was assessed under arthroscopic visualisation at the time of surgery to an accuracy of ±1 cm².

MAT failure was defined as conversion to arthroplasty or revision MAT. ACI failure was defined as requiring revision ACI, a similar cartilage restoration procedure, or conversion to arthroplasty. Overall failure was defined as failure of MAT, ACI, or both. To acquire these data, patients were asked whether conversion to arthroplasty, revision MAT/ACI/similar cartilage restoration procedure, or (sub)total resection of meniscal graft tissue had occurred as part of the standard data acquisition process. Patients who had undergone such procedures at our institution were identified through our internal database.

For postoperative MRI evaluation, the senior author assessed postoperative MRI scans performed during routine follow-up visits at an average of 40.4 ± 20.1 months after index surgery. The Magnetic Resonance Observation of Cartilage Repair Tissue (MOCART) score [36] was used to evaluate cartilage repair tissue quality. Note that MOCART scores were calculated based on low-field-strength MRI scans (Esaote Biomedica DE GmbH, O-Scan, 0.31 T).

Surgical technique

The senior author performed all surgeries using a minimally invasive arthroscopic approach. In brief, a comprehensive preoperative examination was performed, including a complete physical examination of the affected extremity, radiographic evaluation, and MRI. Fresh-frozen meniscal allografts (RTI Surgical Inc.) were sized according to MRI scans as described elsewhere [37, 38] and fixed using a bridge-in-slot technique. Arthroscopic bridge-in-slot fixation was used for both medial and lateral MAT (cf. Figures 2 and 3). To introduce the meniscal allograft including the bone-bridge into the joint, the corresponding arthroscopic portal was slightly enlarged (cf. Figure 2).

Figure 2

Medial and lateral meniscal allograft prior to implantation. Intraoperative photograph depicting prepared medial (a) and lateral (b) meniscal allografts with bone bridge just prior to implantation.

Figure 3

Key surgical steps. Intraoperative arthroscopic view depicting key surgical steps: (a) bone trough with shuttle sutures, (b) outside-in suturing of the pars intermedia, and (c) meniscal allograft in situ.

Cartilage for ACI was harvested during the MAT procedure and sent to the lab (CO.DON GmbH) for expansion. Arthroscopic ACI was typically performed approximately 6–8 weeks after MAT surgery. Two patients received ACI prior to MAT due to graft availability and conversion from meniscal repair to MAT, respectively. Two patients had cartilage harvested first and then underwent simultaneous MAT and ACI. A detailed description of the ACI surgical technique has been published elsewhere [39].

Outcomes

The primary outcome of interest was the pre- to postoperative change in IKDC scores. Secondary outcome measures included pre- to postoperative changes in KOOS, Tegner, and Lysholm scores, as well as patient satisfaction, failure rates, reoperation rate, and MOCART scores.

Statistical analysis

The arithmetic mean, standard deviation (SD), median, and range were calculated as appropriate. Data normality was assessed using Shapiro–Wilk tests, Q–Q plots, and histograms. Nonparametric Wilcoxon signed-rank tests were applied to determine whether the differences in pre- and postoperative outcome parameters were significant. Subgroup analyses were performed using the Mann–Whitney U test. Incomplete data sets were excluded from the corresponding analyses. Due to the explorative design of this study, no alpha adjustment was used. All tests were two-sided and a p-value of ≤0.05 was considered significant. There was no prospective calculation of study size because of the expected very low numbers of combined cases. Data analysis was performed using IBM SPSS for Windows (version 27).

RESULTS

Out of 27 eligible patients, 7 (26%) were lost to follow-up or lacked follow-up data of a minimum of 12 months postoperatively. The final study population included 20 patients (seven female and 13 male), with an average follow-up time of 72.6 ± 34.4 months after index surgery. The average age at surgery was 31.9 ± 9.7 years. The lateral compartment was affected in 15 cases, compared to five patients with medial compartment pathology. The average total chondral defect size was 4.6 ± 3.6 cm². The average number of prior surgical procedures affecting the index knee joint was 1.6 ± 1.0. Prior surgical procedures are reported in detail in Table 2. The anterior cruciate ligament (ACL) was intact in all patients, with three patients having undergone ACL reconstruction prior to index surgery. Baseline patient characteristics are summarised in Table 3. Complete datasets were available for all outcome measures, except VAS for overall satisfaction (n = 17) and MOCART score (n = 14). No symptomatic extrusion of meniscal allograft tissue was observed.

Table 2. Prior surgical procedures. Detailed report of prior surgical procedures of each patient in our cohort.

ID	Number of prior surgical procedures	Types of surgery performed
8	2	1) Transfer of discoid meniscus to lateral reconstruction 2) ACI
12	2	1) Partial resection of lateral meniscus 2) Meniscectomy
14	3	1) ACL reconstruction 2) Meniscectomy 3) ACL revision
18	1	1) Tibial plateau fracture
21	2	1) Meniscectomy, MFx lateral femoral condyle/tibial plateau 2) Meniscectomy
23	1	1) Meniscectomy, chondroplasty lateral femoral condyle
28	1	1) Lateral meniscus transversion
31	1	1) Meniscectomy, AMIC lateral femoral condyle
34	2	1) Meniscectomy 2) Suturing of medial meniscus, Pridie drilling
37	2	1) Meniscectomy 2) Meniscectomy
25	2	1) Meniscectomy, ACL reconstruction 2) Medial MAT, lateral synthetic meniscus implantation
9	2	1) Meniscectomy 2) AMIC medial tibial plateau
36	4	1) ACL reconstruction 2) Meniscectomy 3) HTO, MFx medial femoral condyle/tibial plateau 4) Arthrolysis, debridement, hardware removal

• Note: Patients who only underwent ipsicompartmental meniscectomy prior to MAT are omitted.
• Abbreviations: ACI, autologous chondrocyte implantation; ACL, anterior cruciate ligament; AMIC, autologous matrix-induced chondrogenesis; HTO, high tibial osteotomy; MAT, meniscal allograft transplantation; MFx, microfracture.

Table 3. Baseline patient characteristics.

Characteristics	Unit of measurement/coding	Range or count (n)	Arithmetic mean ± SD or proportion
Age at index surgery	Years	15–47	31.9 ± 9.7
Sex	Male	13	65%
	Female	7	35%
BMI	kg/m^2	19–28	24.1 ± 2.3
Index side	Left	6	30%
	Right	14	70%
Affected compartment	Lateral	15	75%
	Medial	5	25%
Defect size	cm^2	1–15	4.6 ± 3.6
Presence of kissing lesion	No	13	65%
	Yes	7	35%
Duration between MAT and ACI surgery	Weeks (negative values indicate ACI prior to MAT)	−35–25	5.4 ± 10.8
Number of prior surgeries		0–4	1.6 ± 1.0
Follow-up	Months	12–135	72.6 ± 34.4

• Abbreviations: ACI, autologous chondrocyte implantation; BMI, body mass index; MAT, meniscal allograft transplantation; SD, standard deviation.

Marked pre- to postoperative improvements were observed across all longitudinal outcome measures. Statistically significant differences between pre- and postoperative values were found across all scores, except KOOS symptoms and VAS for patient satisfaction (Table 4). IKDC scores improved from 52.1 ± 16.9 to 68.5 ± 16.3 points. The average MOCART score was 68.9 ± 16.8 at an average follow-up of 40.4 ± 20.1 months.

Table 4. Univariate analysis of outcome measures by time point (pre- versus postoperative) compared using nonparametric Wilcoxon tests, as well as deltas (postoperative – preoperative values).

Outcome measure	Unit of measurement/coding	Arithmetic mean ± SD or median (range)			p value
		Preoperative	Postoperative	Delta
IKDC	%	52.1 ± 16.9	68.5 ± 16.3	16.3 ± 23.9	0.003
Lysholm	%	61.5 ± 21.7	78.5 ± 12.9	17.0 ± 21.6	0.004
Tegner	0–10 points	3.5 (1–4)	4.0 (2–6)	1.0 (−1.0–5.0)	0.014
KOOS
Symptoms	%	60.0 ± 26.4	76.1 ± 15.2	16.1 ± 30.5	n.s.
Pain	%	62.5 ± 22.5	85.7 ± 12.7	23.2 ± 26.8	0.001
ADL	%	71.6 ± 23.3	90.4 ± 12.6	18.7 ± 25.3	0.003
Sports	%	40.5 ± 29.2	63.8 ± 29.8	23.3 ± 38.6	0.011
QoL	%	32.2 ± 14.5	54.7 ± 24.1	22.5 ± 25.4	0.001
Total	%	59.5 ± 21.4	80.9 ± 13.5	21.4 ± 25.5	0.001
VAS
Satisfaction with affected knee joint	0 (worst) – 10 (best)	4.0 ± 2.9	5.8 ± 2.9	1.8 ± 4.4	n.s.
Satisfaction overall (n = 17)	0 (worst) – 10 (best)	6.1 ± 2.9	7.9 ± 1.4	1.8 ± 3.4	n.s.

• Abbreviations: ADL, activities of daily living; IKDC, International Knee Documentation Committee subjective knee form; KOOS, Knee Injury and Osteoarthritis Outcome Score; QoL, quality of life; SD, standard deviation; VAS, visual analogue scale.

The combined procedure failed in three cases (15%). Of those cases, two were due to the MAT component of the combined procedure (both patients received revision MAT). Only a single ACI failure was observed, with the patient requiring revision ACI. Of the MAT failures, one patient had a kissing lesion, and the other did not. The single ACI failure had a kissing lesion. We observed no statistically significant influence of the presence of a kissing lesion on outcome measures. Details about failed cases are available in Table 5. Representative examples of MRI findings after successful and failed MAT and ACI are depicted in Figures 4 and 5, respectively.

Table 5. Summary of subsequent surgeries by patient.

ID	Number of subsequent procedures	Types of surgery performed
2	1	1) Debridement and resuturing of meniscal allograft, MFx MFC
4	3	1) Resuturing of meniscal allograft 2) Partial resection and resuturing of meniscal allograft 3) Revision MAT
5	2	1) Refixation and resuturing of meniscal allograft 2) Resuturing of meniscal allograft
6	2	1) Resuturing of bucket-handle tear of meniscal allograft 2) Revision MAT
18	1	1) Partial resection and resuturing of meniscal allograft, AMIC, MFx
23	3	1) Partial resection of meniscal allograft, ganglion resection, capsule suturing 2) Cartilage harvesting for revision ACI, partial resection of meniscal allograft 3) Revision ACI, resuturing of meniscal allograft
25	2	1) Debridement, lavage, synovectomy 2) Debridement, lavage
9	1	1) Arthrolysis, debridement

• Note: MFx (intercondylar notch) and AMIC (bone-bridge) were performed in some cases to improve meniscal healing, not as cartilage procedures.
• Abbreviations: ACI, autologous chondrocyte implantation; AMIC, autologous matrix-induced chondrogenesis; MAT, meniscal allograft transplantation; MFC, medial femoral condyle; MFx, microfracture.

Figure 4

Magnetic resonance imaging findings after successful MAT and ACI. The figure depicts postoperative MRI morphology after successful lateral MAT and ACI at 57 days (a, b) and approximately 62 months (c) after MAT. ACI, autologous chondrocyte implantation; MAT, meniscal allograft transplantation.

Figure 5

Magnetic resonance imaging findings after failed MAT and ACI. The figure depicts postoperative MRI morphology after failed lateral MAT and ACI (failure due to MAT) at approximately 77 months post index surgery. ACI, autologous chondrocyte implantation; MAT, meniscal allograft transplantation.

Overall, eight patients (40%) underwent at least one relevant subsequent surgical procedure on the affected knee joint (not including procedures performed during the planned second part of the combined procedure), with an average number of subsequent procedures of 1.9 ± 0.8. Out of 15 total subsequent procedures, 7 (46.7%) involved MAT refixation or resuturing, with only a single reoperation being directly attributable to ACI (6.7%). The number of subsequent surgeries and type of surgery performed are detailed in Table 5.

DISCUSSION

The most important finding of the present study, investigating the effects of combined MAT and ACI in patients with meniscal insufficiency and full-thickness chondral defects, was substantial improvements in IKDC scores, exceeding the minimal clinically important difference, at medium-term follow-up. Other central findings included: (i) successful integration of ACI as evidenced by postoperative MRI, (ii) a low failure rate (15%), and (iii) a high arthroscopic reoperation rate (40%).

Clinical outcome measures exhibited significant enhancements in line with previous studies [18-22, 40-44]. In terms of the primary outcome measure, the findings in the present study align with the results reported by Yoon et al. [22], who observed a 10.2-point average pre- to postoperative improvement in IKDC scores, compared to an improvement of 16.3 ± 23.9 points in the present cohort, exceeding the minimal clinically important difference of 9.8 points [45]. Lysholm and Tegner scores were similarly aligned with previous research [18, 19, 22, 43]. However, Tegner scores only showed small improvements, illustrating that this combined approach should be considered a salvage procedure and return to sports cannot be considered a primary goal.

Failure rate reporting in previous research is highly heterogeneous, making it difficult to draw robust comparisons. However, the overall failure rate in the present study (15%) compares favourably with the relevant literature reporting failure rates of 11.1%–52.6% [19-22]. However, differences in failure criteria must be considered. Two failures in the present cohort were attributable to the MAT component of the combined procedure, indicating MAT as the limiting factor in this combined approach, as opposed to ACI, with only one ACI-related failure.

In alignment with previous research, a high arthroscopic reoperation rate of 40% was observed, suggesting a considerable risk of reoperation [19-22, 43]. Notably, out of 15 total subsequent procedures across eight patients, nine reoperations (60%) were directly attributable to MAT, with only a single reoperation directly attributable to the ACI component of the procedure (not including cartilage harvesting), underlining MAT as the limiting factor. This comparatively high arthroscopic reoperation rate for MAT can be considered acceptable, bearing in mind the lack of alternative treatment options and reoperations mostly comprising comparatively minor arthroscopic procedures. However, it is crucial to communicate this considerable risk of reoperation to patients to align expectations. Average MOCART scores in this cohort suggested satisfactory cartilage regeneration after combined MAT and ACI, aligning with [46], a phase 2 dose-confirmation trial of isolated ACI.

While existing literature supports significant postoperative improvements after combined MAT and ACI, the comparison with results after isolated MAT or ACI is not as clear-cut, with some studies reporting inferior results after combined surgery [19, 22], whereas others found no significant differences [18, 41, 43, 44]. The current study, while showing significant improvements, did not match those reported after isolated MAT in recent meta-analyses [9, 11], suggesting slightly inferior results after combined surgery. However, the discrepancies could also be attributed to more complex pathologies and worse preoperative joint status in patients receiving combined surgery. Interestingly, the failure rate in this cohort was closely aligned to graft survivorship after isolated MAT as reported in a recent review [9]. However, considering the reported decline in postoperative clinical outcome measures over time, the long-term success of this procedure needs further exploration. Overall, these findings indicate that while improvements in clinical outcome measures after combined surgery might be less than those after isolated MAT, combined MAT and ACI are able to achieve statistically significant and clinically meaningful improvements in several clinical outcome measures and provide good patient satisfaction, with failure rates comparable to those after isolated MAT.

This study has several important limitations, warranting discussion, including its observational design and small sample size, which could introduce biases and limit the generalisability of the findings. Furthermore, our results provide a medium-term average follow-up, with a relatively short minimum follow-up duration of 12 months. This compromise was made to increase the number of eligible patients, as combined MAT and ACI is a rare procedure. Another limitation is the lack of a control group. However, it was not feasible to include one, given the already limited treatment group. Despite these limitations, the study contributes valuable insights to the emerging body of research on combined MAT and ACI, guiding clinical decision-making when encountering the challenging scenario of a young patient with loss of meniscal function and concomitant full-thickness cartilage damage, where biologic treatment options are severely limited, demonstrating an effective alternative to arthroplasty.

CONCLUSIONS

Combined arthroscopic MAT and ACI is a viable, safe, and effective treatment approach for younger patients with meniscal insufficiency and coexisting full-thickness cartilage damage, where alternative treatment options are limited. The procedure achieved significant improvements in clinical outcome measures and patient satisfaction with low failure and high arthroscopic reoperation rates. MAT was the limiting part of this combined procedure, with most failures and reoperations being attributable to MAT, as opposed to ACI.

AUTHOR CONTRIBUTIONS

Johannes Pawelczyk and Rainer Siebold conceptualised and designed the study. Johannes Pawelczyk carried out data collection, manuscript preparation, and manuscript revision. Ilias Fanourgiakis and Sven Feil were involved in drafting and revising the manuscript, as well as evaluating the clinical outcome measures. Rainer Siebold led the design and conceptualisation of the study, performed the surgical procedures, evaluated postoperative MRI scans, and critically revised the manuscript. All authors reviewed and approved the final draft of the manuscript.

ACKNOWLEDGEMENTS

The authors thank Prof. Dr. phil. Sven Schneider, MA, for his expert handling of the statistical analysis. His contributions were critical to the success of this research. The authors have no funding to report.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflict of interest.

ETHICS STATEMENT

ATOS Clinic Heidelberg, Ethics Committee, Bismarckstraße 10-15, 69115 Heidelberg, Germany. Chairman: Prof. Dr. med. Christoph Becher. Reference number: 12021. Written informed consent was obtained from all patients prior to enrolment in the study.