Transtendon repair vs. tear completion in partial-thickness rotator cuff tears: systematic review and meta-analysis

Abstract

Background

Partial-thickness rotator cuff tears (PTRCTs) are common and present a treatment dilemma between preserving the intact tendon with transtendon repair or converting the lesion to a full-thickness tear before repair. Comparative evidence on clinical outcomes and retear rates lacks consensus regarding the optimal treatment of PTRCTs. Therefore, we aimed to systematically compare functional outcomes and tendon integrity following transtendon vs. tear completion repair for PTRCTs.

Methods

A systematic search of PubMed and Scopus was conducted from inception to May 2025. Studies directly comparing the 2 surgical approaches and reporting post-operative functional scores (Constant–Murley, American Shoulder and Elbow Surgeons [ASES], visual analog scale) or imaging-confirmed tendon integrity were included. Pooled mean differences (MDs) and odds ratios were calculated using random-effects models. Risk of bias was assessed with the Newcastle-Ottawa Scale and the Cochrane risk of bias tool.

Results

Nine studies (3 randomized trials, 6 observational; 719 patients—336 transtendon, 385 tear completion repairs) were included. Pooled analysis demonstrated no significant differences between the techniques for Constant–Murley (MD: −0.79, 95% confidence interval [CI]: −1.71 to 0.12; P = .09), ASES (MD: −1.04, 95% CI: −3.90 to 1.83; P = .48), or visual analog scale (MD: 0.11, 95% CI: −0.03 to 0.26; P = .11). Retear rates were also similar (odds ratio: 0.97, 95% CI: 0.39-2.44; P = .96). Subgroup analysis showed no advantage for either technique in articular-sided tears in Constant score (MD: −0.79; 95% CI: −2.47 to 0.88; P = .56) and ASES score (MD: 1.2; 95% CI: −1.21 to 3.61). For bursal-sided tears, tear completion yielded higher Constant–Murley scores (MD: −1.08; 95% CI: −1.88 to −0.28; P = .007).

Discussion

This meta-analysis found no clinically meaningful difference between transtendon and tear completion repair in terms of shoulder function or tendon healing. Surgical choice should be individualized according to tear characteristics, intraoperative findings, and surgeon expertise.

Partial-thickness rotator cuff tears (PTRCTs) are a common cause of musculoskeletal pain and disability, with a reported prevalence of 13%-32%.²⁵ The pathophysiology of the disease is closely linked to increasing age, age-related changes in metabolism, and decreased vascularity of the tendons, which contribute to degeneration and susceptibility to tears of the rotator cuff. Extrinsic factors include shoulder trauma, repetitive overhead activity (particularly athletes), subacromial impingement, and shoulder instability.²⁵ These injuries are associated with variable clinical presentation from asymptomatic PTRCTs to reduced range of motion, crepitus, and nocturnal pain.^12,25 Advances in imaging modalities (ie, magnetic resonance imaging [MRI] and ultrasonography) have facilitated early diagnosis of PTRCT, while the optimal treatment of these injuries is still under investigation.^25,28

Treatment should be tailored based on patients' age, activity level, tear onset, and etiology. Although conservative treatment is usually initially preferred, spontaneous healing of the tendons rarely occurs, and tear progression is highly likely.³⁵ Thus, surgical treatment is often necessary. The 2 most common arthroscopic repair techniques for PTRCTs include the transtendon repair and repair after full tear completion. The transtendon (or in situ) repair technique has the advantage of preserving the intact part of the injured tendon, which allows for more anatomic footprint reconstruction and an adequate length–tension relationship.² The technique achieves excellent long-term results regarding pain, functionality, and return-to-sports time.^29,30 Despite favorable results, it is technically demanding, and some patients may experience mild residual symptoms.⁴ On the other hand, repair after conversion to a full-thickness tear allows for the use of routine rotator cuff repair techniques, and the removal of degenerative tissue promotes healing and mitigates pain.^2,25 Although it exhibits satisfactory clinical and anatomical results,^19,21 it requires compromise of the intact tendon, which raises concerns regarding tendon integrity.³⁶

Given the competing advantages and limitations of both techniques, consensus on the optimal surgical approach remains lacking. Previous comparative studies and meta-analyses have been limited by small sample sizes, heterogeneous outcomes, and a focus on articular-sided tears. To provide a more comprehensive evidence synthesis, this systematic review and meta-analysis aimed to compare functional outcomes and tendon integrity following transtendon repair vs. tear completion repair in patients with PTRCTs, including both articular- and bursal-sided tears.

Methods

This review followed the Cochrane guidance¹⁷ and the Preferred Reporting Items for Systematic reviews and Meta-Analyses 2020 guidelines.²⁶ The protocol was registered in PROSPERO (CRD420251112977; registered 2025).

Search strategy

A systematic search of Medline (via PubMed) and Scopus databases was conducted from inception to May 2025. Search terms combined concepts related to the pathology (rotator cuff tear, partial thickness, PASTA, articular-sided, bursal-sided) and the interventions (transtendon, transtendinous, in situ repair, tear completion repair, full-thickness conversion). Only English-language publications were considered. An initial screening of titles and abstracts was performed independently by 2 reviewers (D.V.P. and A.K.), excluding irrelevant articles. Potentially eligible articles were read in full and critically appraised for inclusion. Reference lists of eligible articles were also screened. Disagreements were resolved by consensus with the assistance of other team members.

Eligibility criteria

Observational studies (retrospective or prospective) and randomized controlled trials (RCTs) comparing the post-operative functional outcomes and rotator cuff integrity following in situ repair vs. tear completion before repair for PTRCT were searched. Eligible studies reported post-operative functional outcomes using standardized scores such as the Constant–Murley score, the American Shoulder and Elbow Surgeons (ASES) score, or the visual analog scale (VAS) score for the 2 groups of patients (ie, patients who underwent in situ repair vs. patients who underwent tear completion before repair). Moreover, studies were considered eligible if they reported information regarding the post-operative rotator cuff integrity based on imaging evaluation (MRI or ultrasound) for the 2 patient groups. For MRI evaluation, retear was defined based on the Sugaya classification as type IV and V, or as “noncontinuity with insufficient thickness of the repaired tendon with heterogeneous low intensity.” For ultrasound evaluation, partial-thickness and full-thickness tears were both defined as “retears.” In order to be included, studies should report sufficient data, including mean values and standard deviations for the functional scores, or post-operative retear rates for both study groups.

Data extraction

Two reviewers (D.V.P and A.K.) independently extracted the following data from each study using a predefined extraction form: first author, publication year, country, study period, design, setting (hospital-based vs. database), side of partial-thickness tear (articular-sided vs. bursal-sided), follow-up, number of participants, mean values and standard deviation for functional scores, retear rates, and imaging modality for evaluation of rotator cuff integrity. Then, 2 types of effect estimates were calculated from the individual studies: (i) mean difference (MD) between the groups for post-operative functional scores and (ii) odds ratio (OR), for post-operative rotator cuff integrity.

Risk of bias assessment

Observational studies were assessed using the Newcastle-Ottawa Scale (NOS), which awards up to 9 stars across 3 domains (selection, comparability, outcome). Studies scoring < 6 stars were considered at high risk of bias (RoB).¹³ RCTs were evaluated with the Cochrane Collaboration RoB tool across 7 domains, classifying studies as low, unclear, or high risk.¹⁶ Two reviewers (D.V.P and A.K.) independently performed all assessments, and disagreements were resolved by consensus.

Quantitative data synthesis

Pooled MDs were calculated for continuous outcomes, and ORs with 95% confidence intervals (CIs) for dichotomous outcomes, using random-effects models (DerSimonian-Laird).⁸ Small-study effects were explored with Begg and Egger tests^1,9 (noting limited power with < 10 studies) and visually with funnel plots. Statistical heterogeneity was investigated with the Cochran Q-test⁶ with a 0.10 level of significance and quantified using the I² statistic,¹⁸ which reports the proportion of variability across studies that is due to between-study heterogeneity rather than chance. The I² statistic was interpreted according to the Cochrane Handbook guide: 0%-40% might not be important; 30%-60% may represent moderate heterogeneity; 50%-90% may represent substantial heterogeneity; while 75%-100% represents considerable heterogeneity. Sensitivity analyses were planned by tear side (articular-sided vs. bursal-sided), follow-up duration (<2 years vs. >2 years), and study design (RCTs vs. observational). All the analyses were performed using Stata 19 software (StataCorp, College Station, TX, USA). A two-tailed P value <.05 was considered statistically significant for all analyses, except for the assessment of heterogeneity (where a level of 0.10 was used).

Results

Studies

The electronic search identified a total of 1,748 articles in PubMed and Scopus databases, while after duplicate removal, 1,677 studies remained available for the initial screening. The titles and abstracts were screened, and 1,541 articles were removed because they were irrelevant to the topic of interest. The full text of the remaining 136 articles underwent a thorough review to evaluate whether they met the inclusion criteria. Following this full-text review, 127 studies were excluded because either they were not evaluating partial rotator cuff tears (n = 12), were not comparing the 2 treatment techniques (n = 101), did not report the information of interest (n = 8), or were laboratory studies (n = 6). Ultimately, 9 studies were included^{3,5,11,22,24,31,32,37,38} (Fig. 1).

Figure 1.

Diagram of the search and selection process (flowchart).

Of these, 3 were RCTs^3,11,31 and 6 were observational studies (1 prospective²² and 5 retrospective^{5,24,32,37,38} (Table I). Three were conducted in Italy,^3,5,11 3 in China,^24,37,38 and 3 in South Korea.^22,31,32 Across studies, 719 patients (721 shoulders) were analyzed, including 336 undergoing transtendon repair and 385 undergoing tear completion repair. Two patients underwent bilateral procedures, accounting for the discrepancy between the number of shoulders and patients.

Table I.

Studies included in the evidence synthesis (n = 9).

Study, publication yr	Country	Time period	Design	Side of PTRCT	Mean follow-up period (mo)	No. of PTRCT	No. of PTRCT in study groups	Functional outcomes In situ vs. tear completion (MD, 95% CI)			Tendon integrity In situ vs. tear completion (OR, 95% CI)
								ASES score	Constant–Murley score	VAS score	Retear rate
Shin et al31	S. Korea	2006-2008	RCT	Articular	31.3	48	In situ: 24 Tear completion: 24	2.90 (1.37, 4.43)	−2.30 (−3.58, −1.02)	0.30 (0.12, 0.48)	NR
Castagna et al3	Italy	2006-2009	RCT	Articular	≥24	74	In situ: 37 Tear completion: 37	NR	−3.90 (−6.64, −1.16)	0.20 (−0.47, 0.87)	0.19 (0.01-4.08)
Franceschi et al10	Italy	2007-2009	RCT	Articular	38	60	In situ: 32 Tear completion: 28	1.00 (−2.67, 4.67)	1.00 (−2.65, 4.65)	NR	0.87 (0.05-14.60)
Shin et al32	S. Korea	2008-2012	Cohort	Bursal	32.5	84	In situ: 47 Tear completion: 37	−4.80 (−6.38, −3.22)	−1.30 (−2.23, −0.37)	0.00 (−0.22, 0.22)	1.05 (0.22-5.03)
Kim et al21	S. Korea	2008-2011	Cohort	Articular and bursal	19.1	92	In situ: 47 Tear completion: 45	−6.50 (−11.59, −1.41)	0.00 (−2.12, 2.12)	0.70 (−0.09, 1.49)	0.24 (0.05-1.23)
Liu et al23	China	2014-2015	Cohort	Articular	13	68	In situ: 30 Tear completion: 38	−3.10 (−7.12, 0.92)	NR	−0.17 (−0.59, 0.25)	NR
Castricini et al5	Italy	2003-2014	Cohort	Articular	73.9	153	In situ: 59 Tear completion: 94	NR	−0.20 (−0.52, 0.12)	NR	2.54 (0.91-7.09)
Zhuo et al37	China	2019-2020	Cohort	Bursal	14.9	58	In situ: 28 Tear completion: 30	−0.47 (−2.31, 1.37)	−0.47 (−2.05, 1.11)	0.08 (−0.08, 0.24)	2.23 (0.19-26.06)
Zhang et al36	China	2017-2020	Cohort	Articular	≥24	84	In situ: 32 Tear completion: 52	2.30 (−0.79, 5.39)	2.50 (−0.50, 5.50)	0.00 (−0.46, 0.46)	NR

RCT, randomized controlled trial; PTRCT, partial-thickness rotator cuff tear; ASES, American Shoulder and Elbow Surgeons; VAS, visual analog scale; OR, odds ratio; CI, confidence interval; NR, not reported.

Baseline demographics were comparable between groups. Mean age was 56.7 years in the transtendon group and 58.0 years in the tear completion group.

Risk of bias assessment

RoB in observational studies (n = 6) was assessed with the NOS tool: 1 study scored 9 stars, 2 scored 8, and 3 scored 7, indicating generally low RoB (Supplementary Table S1). The RCTs (n = 3) were appraised with the Cochrane Collaboration RoB tool. All were judged to have some methodological concerns, primarily due to unclear allocation concealment and lack of blinding (Supplementary Table S2).

Functional outcomes and retear rate

Eight studies reported post-operative Constant–Murley scores and 6 reported ASES scores (Table II). Pooled analysis showed no significant difference between transtendon and tear completion repair: Constant–Murley (MD: −0.79, 95% CI: −1.71 to 0.12; P = .09; Fig. 2) and ASES (MD: −1.04, 95% CI: −3.90 to 1.83; P = .48; Fig. 3). Seven studies reported VAS scores. The pooled MD was 0.11 (95% CI: −0.03 to 0.26; P = .11; Fig. 4). The tests for small-study effects were not statistically significant (Begg's P = .71, .36, 0.54; Egger P = .57, .42, 0.94 for Constant–Murley, ASES, and VAS, respectively); however, their power was limited due to the small number of studies included. Funnel plots suggested slight asymmetry for Constant–Murley and ASES, but not for VAS score (Supplementary Figures S1–S3). Heterogeneity was substantial for Constant–Murley (I² = 71.1%), substantial/considerable for ASES (I² = 89.6%), and moderate for VAS (I² = 37.0%) (Table II).

Table II.

Meta-analysis results.

Outcome measure	No. of studies	Random-effects estimates		Tests of homogeneity			Tests of publication bias
		MD (95% CI)	OR (95% CI)	Q value (d.f.)	P value	I2	Begg P value	Egger P value
Constant–Murley score	8	−0.79 (−1.71 to 0.12)	-	24.21 (7)	.001	71.1%	.71	.57
ASES score	6	−1.04 (−3.90 to 1.83)	-	57.91 (5)	<.001	89.6%	.36	.42
VAS score	7	0.11 (−0.03 to 0.26)	-	9.53 (6)	.14	37.0%	.54	.94
Retear rate	6	-	0.97 (0.39-2.44)	7.46 (5)	.19	33.0%	.25	.31

MD, mean difference; OR, odds ratio; CI, confidence interval; d.f., degrees of freedom; ASES, American Shoulder and Elbow Surgeons; VAS, visual analog scale.

Figure 2.

Forest plot of individual studies, and meta-analysis, examining the difference in post-operative Constant–Murley scores between transtendon repair and tear completion techniques, in partial-thickness rotator cuff tears. CI, confidence interval; SD, standard deviation.

Figure 3.

Forest plot of individual studies, and meta-analysis, examining the difference in post-operative ASES scores between transtendon repair and tear completion techniques, in partial-thickness rotator cuff tears. ASES, American Shoulder and Elbow Surgeons; CI, confidence interval.

Figure 4.

Forest plot of individual studies, and meta-analysis, examining the difference in post-operative VAS scores between transtendon repair and tear completion techniques, in partial-thickness rotator cuff tears. VAS, visual analog scale; CI, confidence interval.

Six studies reported post-operative tendon integrity. Pooled analysis demonstrated no significant difference regarding the retear rates between transtendon and tear completion repair (OR: 0.97, 95% CI: 0.39-2.44; P = .96; Fig. 5). Heterogeneity was moderate (I² = 33%; Table II), and no clear evidence of small-study effects was observed: The P values for the Begg and Egger tests were .25 and .31, respectively, while the funnel plot did not indicate any asymmetry (Supplementary Figure S4).

Figure 5.

Forest plot of individual studies, and meta-analysis, comparing retear risk following transtendon repair vs. tear completion techniques, in partial-thickness rotator cuff tears. CI, confidence interval.

Subgroup analyses

We conducted a subgroup analysis for studies evaluating articular-sided and bursal-sided PTRCT separately. The Constant–Murley score was compared between patients undergoing transtendon repair and those undergoing tear completion before repair. For articular-sided PTRCT (5 studies), the 2 techniques resulted in comparable scores (MD: −0.79; 95% CI: −2.47 to 0.88; P = .56). In contrast, for bursal-sided PTRCT (2 studies), transtendon repair was associated with a lower Constant–Murley score (MD: −1.08; 95% CI: −1.88 to −0.28; P = .007). The ASES score was assessed in 4 studies for articular-sided PTRCT and in 2 for bursal-sided PTRCT. Scores were comparable between the 2 techniques in both subgroups (MD: 1.2; 95% CI: −1.21 to 3.61, and MD: −2.6; 95% CI: −6.90 to 1.58, respectively). Finally, the VAS score was evaluated in 4 studies for articular-sided PTRCT and 2 studies for bursal-sided PTRCT, showing no significant differences between treatment techniques (MD: 0.12; 95% CI: −0.11 to 0.37; P = .36, and MD: 0.05; 95% CI: −0.07 to 0.18; P = .78, respectively). The retear rate was evaluated in 3 studies for articular-sided PTRCT. There was no difference regarding the retear rate between the 2 techniques (OR: 1.33, 95% CI: 0.33-5.29, P = .67). Subgroup analysis based on type of studies (RCT vs. observational studies) showed no differences between the 2 techniques for either type of studies (P > .05). Similarly, subgroup analysis based on duration of follow-up showed that there is no difference between the 2 techniques when evaluating only studies with a follow-up longer than 2 years or studies with a shorter follow-up period (P > .05).

Discussion

The results of this meta-analysis indicate that both repair techniques yield comparable outcomes in terms of functionality, pain, and post-operative complications in PTRCT patients. Interestingly, although there was no significant difference in the functional scores for the articular-sided PTRCTs, the subgroup analysis revealed a MD of 1.08 in the Constant–Murley score for bursal-sided PTRCTs, favoring tear completion repair. However, this difference is much lower than the Minimal Clinically Important Difference for the Constant–Murley score, which is 10 points.²³ Therefore, the reported difference can be considered clinically not important.

The 2 arthroscopic repair techniques for PTRCTs include the transtendon repair and repair after full tear completion. Transtendon repair preserves the uninjured part of the tendon, which may lead to decreased tendon retraction and a more optimal length–tension relationship.⁴ However, aside from the theoretical advantages, the procedure is technically demanding and surgical expertise is necessary. On the contrary, repair after full tear completion is technically simpler, as it allows for routine repair techniques. However, converting a PTRCT into a full-thickness tear compromises the intact fibrocartilage connection and, to some extent, replaces healthy tissue with fibrous scar tissue, raising concerns about retear susceptibility.¹⁰

Although there have been previous meta-analyses comparing transtendon and completion repair, ours is to date the most comprehensive on this topic, as it includes all types of PTRCTs, including bursal and articular-sided tears. The results of our study align with the existing literature, reporting no significant difference between the 2 techniques in post-operative Constant–Murley, ASES, and VAS pain scores, as well as post-operative complications, ie, retear rates. Other recent meta-analyses, such as those by Yang et al and by Thamrongskulsiri et al reported similar results regarding articular-sided PTRCTs. Specifically, no significant difference was observed between transtendon and completion repair.^34,36 These studies restricted their analyses only to articular-sided PTRCTs, whereas our study provides a comprehensive synthesis including both articular- and bursal-sided tears. This broader scope enhances the generalizability and clinical applicability of our findings. Moreover, our meta-analysis incorporated the most recently published comparative studies, such as the one by Zhang et al,³⁷ which increased statistical power and precision. For example, our pooled analyses regarding articular-sided PTRCT included 5 studies for Constant–Murley scores, 4 for ASES scores, and 4 for VAS scores, representing a more complete evidence base than the previous meta-analyses, which relied only on 2 to 3 studies per outcome. Conversely, Sun et al³³ reported a lower retear rate in patients with articular-sided PTRCTs who underwent transtendon repair compared to those repaired after tear completion (retear rates: 0.043 vs. 0.113, P < .005, respectively). However, of the studies included in this meta-analysis, only 2 RCTs directly compared the 2 procedures,^11,31 while the other studies assessed either one of the 2 techniques.³³ Moreover, in their systematic review, Jordan et al²⁰ suggested higher rates of early shoulder stiffness in patients subjected to transtendon repair, although these results were drawn largely on case-series studies, and no meta-analysis was conducted. In another recent meta-analysis, Dalmas et al⁷ focused exclusively on bursal-sided PTRCTs and included studies evaluating either technique in isolation rather than studies directly comparing transtendon and tear completion repair. Including noncomparative studies can introduce bias when synthesizing effect estimates between 2 interventions, which may compromise the robustness of conclusions. In contrast, our meta-analysis strictly included only head-to-head comparative studies, thereby ensuring higher internal validity of the pooled results.

The main downside of transtendon repair is the presence of residual symptoms, reported in up to 40% of cases, possibly reflecting poor tendon quality.⁴ Residual symptoms usually include shoulder discomfort at the extremes of abduction and internal rotation. According to Castagna et al,⁴ patients with residual symptoms scored lower in University of California-Los Angeles, Simple Shoulder Test, and Constant scores compared to those without discomfort, although this difference was insignificant. Predictive factors for such symptoms included a smaller footprint exposure, larger tendon retraction, and elderly patients with atraumatic tears.⁴ Therefore, when these patient- and tear-related characteristics are observed, completion repair may be preferable.

Biomechanical and histological comparative studies of the 2 techniques reach conflicting conclusions. In their cadaveric study, Gonzales et al¹⁵ report superior biomechanical properties for in situ repair of partial, articular-sided supraspinatus tears. Specifically, shoulders subjected to transtendon repair exhibited less gap formation during rotational movement and a higher load-to-failure threshold compared to shoulders subjected to completion repair.¹⁵ This could be attributed to the preserved tendon distributing the axial and torque loads on the suture anchors more evenly.¹⁵ In addition, transtendon repair facilitates a more anatomic footprint reconstruction with minimal initial gap formation, both of which contribute to tendon integrity and resilience to retears.¹⁵ In contrast, a biomechanical and histological study on an animal model reported a higher load-to-failure in rabbit shoulders that underwent completion repair, and an increased rate of parallel-oriented collagen fibers microscopically compared to transtendon repair.²⁷ To support these findings, Gereli et al¹⁴ microscopically examined rat shoulders with bursal-sided PTRCT after completion or in situ repair. At ten days after surgery, specimens from the shoulders subjected to completion repair exhibited increased neovascularization and fibroblast diameter, while at 30 days post-operatively there was increased collagen 1 alpha 1 concentration and fibroblast count, compared to shoulders subjected to transtendon repair.¹⁴ Therefore, completion repair promotes an enhanced healing process due to the removal of devitalized tissue through débridement.¹⁴ Finally, tumor necrosis factor-α levels were comparable between the 2 techniques, indicating similar inflammation levels, although completion repair is more invasive.¹⁴ In the end, this healing process initiated after débridement and completion repair may not equal superior biomechanical features.¹⁴

Our study has limitations that should be considered. First, substantial heterogeneity was observed among studies reporting ASES and Constant scores, which may limit the generalizability of our findings. In addition, our meta-analysis included studies with diverse designs, including RCTs as well as prospective and retrospective observational studies, each of which is subject to distinct sources of bias. Nonetheless, the high methodological quality of the observational studies, as assessed by the NOS, and the low likelihood of publication bias among the RCTs support the overall reliability of our findings. Furthermore, there was no standardized follow-up period across studies, which ranged from 13²⁴ to 74 months,⁵ and diagnostic approaches varied (arthroscopy vs. MRI for initial diagnosis), with inconsistent tear classifications. Despite these limitations, this meta-analysis represents, to date, the most comprehensive synthesis of the literature comparing transtendon repair and tear completion for PTRCTs, encompassing both articular- and bursal-sided lesions. Subgroup analyses allowed tear-type insights, enhancing the clinical applicability of our results.

Based on the findings of this meta-analysis, transtendon repair and repair following tear completion yield comparable outcomes in terms of shoulder function, pain relief, improvements in active range of motion, and retear rates. No statistically significant differences were observed across any of these measures. Despite theoretical differences in tendon histology and biomechanics, our results demonstrate no significant difference in clinical outcomes between the 2 techniques. Consequently, treatment decisions should be guided primarily by individual patient needs, with consideration of the surgeon's expertise and judgment.

Disclaimers:

Funding: The authors received no funds or grants from any sponsor.

Conflicts of interest: The authors, their immediate families, and any research foundations with which they are affiliated have not received any financial payments or other benefits from any commercial entity related to the subject of this article.