Definition of the terms “acute” and “traumatic” in rotator cuff injuries: a systematic review and call for standardization in nomenclature

This systematic review was structured according to the PRISMA Checklist [9]. A literature search was performed focusing on studies investigating acute and traumatic rotator cuff tears. Prospective, retrospective, cohort and case–control studies as well as case series written in English and German were included. Systematic reviews, cadaveric or laboratory studies and studies on non-traumatic or non-acute rotator cuff tears were excluded.

Inclusion or exclusion of studies was based on the aforementioned criteria. Two reviewers (blinded for review) independently analysed titles and abstracts of all identified publications and the complete text of eligible articles. Divergences were resolved in collaboration with a third reviewer (blinded for review).

The following data were abstracted from all included studies: study design, level of evidence, type of imaging used to detect the RCT, time frames used for definition of an acute injury and the definition of the term “traumatic” in the context of an RCT. Level of evidence was assigned according to the widely accepted common grading of Evidence Levels for Primary Research [10, 11]. The methodological quality of the included studies was assessed using the Methodological Index for Non-Randomized Studies (MINORS) [12]. The MINORS checklist consists of 12 items which are scored 0 (not reported), 1 (reported but inadequate) and 2 (reported and adequate) each with a maximum score of 16 for non-comparative studies (nc) and 24 for comparative studies (c). The methodological quality was categorized a priori as follows: a total score of 0–8 or 0–12 was considered as poor quality, 9–12 or 13–18 was considered as fair quality and 13–16 or 19–24 was considered as excellent quality, for non-comparative studies and comparative studies, respectively. Since this review aimed to analyze the definitions of the terms “acute” and “traumatic” for RCTs regardless of the patient outcome, MINORS could only evaluate the quality of the study but could not provide information on the quality of the definitions of the aforementioned terms.

According to the 18 articles defining the term “acute” with regard to RCTs, the time range was defined between less than 2 weeks up to 6 months (Table 1) [39‐56]. These studies can be further subdivided into ten studies, using a period of a minimum of two up to a maximum of 6 weeks as time span for an acute RCT [41, 43, 45‐49, 51‐53] and eight studies determining a time span from 2 to 6 months for their definition of “acute” [21, 39, 40, 44, 50, 54‐56].

Ten studies contained specific explanations of their definition of the term “acute” [41, 43‐47, 52, 53, 55, 56]. Studies performed by Basset and Cofield [41] and Farin and Jaroma [45] found lower degrees of retraction of the tendon and less scarring of the joint within the first 3 weeks after trauma and, therefore, proposed that surgical repair would be easier in this time frame. Therefore, the authors suggest, that 3 weeks should be considered as the time span for the definition of an acute RCT [41, 45]. Two studies further recommended surgery in a period of less than 3 weeks, referring to the study of Basset and Cofield [41, 46, 47]. Three articles set their definition of an acute RCT within 6–8 weeks after injury, using the average time span from several other clinical studies as a rationale for their definition [43, 53, 56]. Duncan et al. [44] argued that the evidence for defining acute and chronic remained unclear, using 6 months as a distinctive threshold. One study referred to the current Swedish national guideline that recommends surgery within 6 weeks [52]. Petersen et al. [55] reported favorable results for patients with RCTs treated within 4 months. The remaining eight studies provided no concise explanation for their definition of an acute RCT [39, 40, 42, 48‐51, 54].

The term “traumatic” was defined in all 46 included studies [3, 5, 13‐56]. The definitions of the term “traumatic” used in each publication are shown in Table 1.

For the definition of a traumatic lesion of the rotator cuff, all included studies required a sudden onset of symptoms following a patient-reported trauma to the shoulder [3, 5, 13‐56].

Furthermore, in 24 studies (52%) only RCTs in previously asymptomatic patients were considered traumatic lesions [13, 15, 17, 19, 21, 25, 31, 35, 38‐40, 42‐44, 46, 48‐56].

The trauma mechanism of the injury was detailed in 23 studies (50%) [3, 13‐18, 20, 21, 24, 27‐32, 34, 38, 39, 46, 48, 54, 55]. Of these, eleven publications (24%) classified injuries with a specific and adequate injury mechanism as traumatic [3, 13, 16, 17, 21, 30, 31, 34, 39, 46, 54], whereas twelve studies (26%) described the trauma mechanism without assessing the respective adequacy for the definition “traumatic” [14, 15, 18, 20, 24, 27‐29, 32, 38, 48, 55]. More extended criteria were used in 14 publications (30%) [3, 16, 19, 21, 23, 39, 42, 44, 46, 47, 50, 52, 53, 55]. One study investigating traumatic tears only included patients younger than 25 years [23]. Loss of function was required in nine studies [19, 21, 39, 42, 44, 46, 52, 53, 55]. The intraoperative macroscopic or radiologic tear morphology was specified in five studies [3, 16, 21, 47, 50]. Furthermore, four publications defined RCTs as non-traumatic, if the patients had intraoperative or radiologic signs of previous shoulder dysfunction [19, 21, 46, 50].

Similar to the biologic repair process of any other tendon tissue, the healing process of rotator cuff tears is time-dependent and commonly divided into three consecutive phases: (1) inflammation, (2) repair/proliferation and (3) remodeling [57‐60]. Each of the phases depends on the interplay of different cell populations, cytokines and growth factors [61]. It is important to note, that the expression of growth factors and cytokines is highly time-dependent.

First of all, immediately after an injury, bleeding leads to the accumulation of cytokines and growth factors such as fibrin, fibronectin, interleukin1-ß (IL1-ß), IL-6, tumor necrosis factor α (TNF-α), insulin-like growth factor-1 (IGF-1), platelet-derived growth factor (PDGF), basic fibroblast growth factor (bFGF) and transforming growth factor ß (TGF)-β. These growth factors peak in expression after 7 days [62, 63] and activate fibroblasts, which play a key role in the activation of the healing process, angiogenesis and remodeling of the delevoping extra-cellular matrix [61, 64, 65]. Furthermore, the release of these proteins results in the attraction of inflammatory cells such as neutrophils, macrophages and platelets, initiating the inflammatory phase, which lasts for approximately 7 days [57‐59, 66].

This initial inflammatory phase is followed by a phase of repair and proliferation, lasting for approximately 2 weeks after the injury [57‐60]. Herein, cell proliferation and angiogenesis are regulated by vasoactive and chemotactic factors such as VEGF and TGF-β, which activate collagen synthesis, angiogenesis and termination of cell proliferation. All of those factors peak 10 days after injury [62, 63]. In this phase, fibroblasts begin to produce collagen, mainly type I and III [57, 58, 62, 63, 67, 68]. While collagen type I is predominant in healthy tendons, the developing scar tissue contains a higher proportion of collagen type III [66].

The subsequent third phase of remodeling begins after a few weeks and can continue for several years [36, 66]. As a part of the remodeling phase, IGF-1 release is predominant 3 weeks after injury and further stimulates the collagen synthesis [58]. As part of this process, a shift to a more physiologic distribution between collagen types I and III is occuring, resulting in an increasingly organized extra-cellular matrix [66, 67]. However, after a tear at the tendon-bone interface, the biological processes are not able to restore the native composition of the structure, resulting in a fibrovascular scar [69]. For example, several studies in animal models have shown that the regenerative tissue formed in supraspinatus tendons in rats is of inferior mechanical and histological quality compared to its native composition [60, 67]. A decrease of all measured growth factors to control-group level was observed after 16 weeks, which could be considered as the biological equivalent to the completion of the healing process [63].

Radiologic imaging is helpful to detect RCTs and to distinguish between acute and chronic tears. Based on the literature, the MRI is considered the standard imaging technique to reliably diagnose an RCT [70‐72]. As MRI is expensive, time-consuming and not ubiquitously available, ultrasound performed by experienced musculoskeletal clinicians is considered a valid imaging tool with reliable results for detecting RCT [73‐75]. In case of an acute and traumatic rotator cuff tear, muscle edema and joint effusion and a wavelike appearance of the central part of the torn tendon, which is also known as the “kinking sign”, can be detected [31]. In contrast, muscular atrophy and fatty infiltration are safe signs for a chronic component of the RCT, which can be visualized considerably more precise by MRI. However, the degree of tendon retraction and thinning of the tendon are unspecific signs and insufficient to distinguish between acute and chronic tears [31]. In the included studies, either ultrasound or MRI was used in 40 of 46 studies (87%) [5, 13‐16, 18‐25, 27, 28, 30, 31, 33‐40, 42‐56]. In the remaining six studies (13%), only X-ray imaging was used or no imaging modality was employed at all [3, 17, 26, 29, 32, 41].

Apart from radiologic imaging, histological investigations may provide information regarding the causality of the RCT. For example, thinning and disorientation of collagen fibers, vascular proliferation, calcification, myxoid degeneration and pronounced chondroid metaplasia can be considered as signs of pre-existing damage or degenerative changes of the rotator cuff tendons [76]. Acute injuries on the other hand, are characterized by hemosiderin appearence, hypervascularization and regenerative changes [77, 78]. While histopathological examinations may facilitate an ex post classification of the RCT type, it is not feasible to exclusively rely on histopathological examinations, when differentiating between acute and chronic RCTs—due to the heterogeneous and contradictory results and general impracticality in the clinical setting [42].

Several factors, including history of trauma and mechanism of injury, contribute to the definition of a traumatic RCT. In terms of injury mechanisms, it is important to define truly traumatic tears more precisely and to distinguish them from RCTs resulting from inadequate injury mechanisms such as contusions, minor impacts or active muscle contractions during weight lifting [79, 80].

From a biomechanical point of view, various injury patterns can lead to excessive eccentric loading and overstretching of the tendons of the rotator cuff resulting in a traumatic RCT [80]. These mechanisms include shearing of the tendons on the glenoid rim, when the maximal tolerated rotation angle is exceeded. For example, a staircase fall with the attempt to prevent falling by holding on the railing results in a passively forced external or internal rotation and abduction with a massive overstretching of anterocranial or posterocranial structures, respectively, possibly leading to a RCT [79, 80]. In addition, axial compression and passive traction can lead to an RCT—a mechanism present in incidents such as a fall on the retroverted arm, a high velocity accident or unprepared catching of a heavy load. Similarly, a shoulder dislocation represents an adequate mechanism capable of causing injury to a healthy rotator cuff tendon [21, 31, 39, 46, 80].

While all included studies reported shoulder trauma as the cause of injury [3, 5, 13‐56], only 24 of the 46 studies (52%) explicitly reported, that the injuries occurred in previously asymptomatic patients [13, 15, 17, 19, 21, 25, 31, 35, 38‐40, 42‐44, 46, 48‐56]. In addition, only eleven studies (24%) described a specific injury mechanism, that could be considered adequate for a traumatic RCT [1, 4, 9, 16, 33, 48, 50, 52, 66, 71, 75].

After reviewing the existing literature, we believe that the terms “acute” and “traumatic” must be considered in combination. The underlying rationale is to exclude so called “acute on chronic”-lesions, where an inadequate trauma leads to a rupture of a pre-damaged rotator cuff tendon [31]. Those tears may not have the same healing capacities as primarily traumatic RCTs. To categorise an RCT as “traumatic”, the patient-reported trauma needs be adequate and follow a biomechanically plausible injury mechanism. There must be a clear differentiation from minor traumatic incidents or inadequate injury mechanisms, indicative of a non-traumatic tear of other etiology. Moreover, for an RCT to be considered traumatic, a previously asymptomatic patient needs to report a sudden onset of symptoms in the shoulder after the trauma.

To be able to detect a truly traumatic and acute RCT, imaging should be performed within the first 2 weeks after injury, as the aforementioned radiologic properties of acute RCT such as muscle edema, joint effusion and the kinking sign can be visualised best in this time interval. In general, RCT should be visualised by ultrasound or MRI for an adequate statement about the ruptured tendon and possible concomitant pathologies. MRI is considered the gold standard imaging modality for RCT, as it allows further assessment of intraarticular structures and, more importantly, enables to detect signs of degenerative changes [70‐72]. Histological investigations do not seem to be mandatory, as no clear evidence could be produced, that this modality is a reliable and practial tool to distinguish between acute and chronic rotator cuff tears.

Nonetheless, it must be taken into account, that due to the very complex interaction of growth factors, cytokines and involved cells, no comprehensive statement about the healing mechanisms and the ideal therapeutic time span for surgery in acute RCT can be made. Yet taking the decrease of most growth factors, cytokines and the collagen shift within the tear zone 8 weeks after trauma into account [58, 59, 63], it must be assumed that the healing capacity is significantly inferior after this period. Therefore, we recommend a time span of no more than 8 weeks to attempt a surgical repair of an acute RCT, to still benefit from superior biological healing capacities posttraumatically. A time period of more than 4 months after trauma appears to be generally unfavorable, due to the dynamics of the aforementioned physiological healing capacities [21, 39, 40, 44, 50, 54‐56].

Our proposed definitions may help standardising the use of the terms “acute” and “traumatic” and, therefore, improve future understanding of the outcomes following rotator cuff repair, but further specification and elucidation of the exact healing processes is required. We are also aware of the inherent limitations implicated in our proposed definitions: First, a significant number of patients does not present within 2 weeks following trauma, making it difficult to detect signs of acute ruptures on ultrasound or MRI. Second, even if the patient presents in time, imaging modalities may not be available in time, especially in rural areas. Third, the exact trauma mechanism is scarcely comprehensible ex post, as the patient may not recall the exact details during the trauma. However, we believe, that it is of critical importance to attain guidelines for the definition of the terms acute and traumatic in the context of rotator cuff tears to be able to compare outcomes of future studies.