Introduction

Shoulder pain is the second most common musculoskeletal disorder and thereby a relevant problem disabling middle-aged and elderly patients in particular [1–5]. In this respect, a meta-analysis summarising 18 studies was recently published by Luime et al. reporting subacromial impingement syndrome (SIS) with a percentage of up to 44 % as one of the most frequently recorded disorders [6]. According to Linsell et al., an even higher prevalence of SIS was found in older patients above 60 years [7]. Although the pathogenesis is not fully understood, most authors describe SIS as a combined result of extrinsic and intrinsic mechanisms, leading to rotator cuff tearing (RCT) and spur formation under the coracoacromial arch [8–11].

Regarding SIS therapy, several authors reported in the past that physiotherapy aimed at strengthening of the muscular motors and stabilisers of the shoulder joint renders satisfactory results especially in patients aged under 60 and represents a cost-effective treatment [12, 13]. In contrast, for those individuals resistant to physical therapy, arthroscopic subacromial decompression (ASD) is the most successful surgical procedure providing clear benefits [1, 11, 14]. Short- and mid-term follow ups of five years after ASD are generally considered advisable [15, 16]. Recent literature comparing conservative therapy and ASD suggests that at least in patients under 60 years equally good clinical results might be achieved [12, 17, 18]. Although physical demands and sports activity are rising notably in the older population, it remains unclear which treatment regimen is superior in isolated SIS [12, 19].

Hence, the aim of this study was to analyse the long-term outcome of patients, having been treated for SIS either conservatively or arthroscopically and to focus on the impact of age at initiation of therapy on the objective and subjective result.

Materials and methods

Study design

This study was designed as a retrospective cohort study with a median follow-up of 55 months. The study was approved by the local ethics committee of the Ludwig-Maximilians-University (reference number: 15/5/2009-GP/AC).

Patient collective

All patients suffering from SI and presenting in our shoulder outpatient clinic were identified retrospectively and enrolled in the study. Patients were identified according to an age range between 20 and 82 years, presence of shoulder pain for more than three and less than six months and meeting the following criteria:

  • pain on abduction of the shoulder with a painful arc

  • positive Neer and Hawkins testing

  • positive impingement test (relief of pain after injection of local anaesthetics into the subacromial space)

Preoperative standard radiographs of the shoulder and an outlet view were performed. Patients with any other pathology such as rotator cuff tear, gleno-humeral instability, cartilage damage (> Outerbridge II), clinically verified acromio-clavicular (AC) joint osteoarthritis, calcifying tendonitis, biceps pathology or signs of cervical root or temporomandibular symptoms verified clinically or during the arthroscopic examination were excluded from the study. Furthermore, patients were excluded if any additional surgical procedures other than subacromial decompression with or without AC joint surgery were performed.

All identified patients were contacted with a letter explaining the purpose of the study, an informed consent sheet for participation in the study and a standardised, validated self-assessment questionnaire (Munich shoulder questionnaire MSQ) along with a pre-stamped return envelope for easy return. If any questionnaire was not returned, the patient was reminded twice. Patients who did not respond at all were excluded from the study.

For assessing the impact of age on the type of treatment of SIS, the patient cohort was divided in two groups according to the median age.

Therapy

Conservative therapy

Patients allocated to conservative treatment received a standardised protocol of physiotherapy, heat, cold pack or/and soft issue treatment (16 sessions with 60 min each for 12 weeks). Then, the patients had active training of the periscapular muscles and strengthening of the stabilising muscles of the shoulder joint at least twice per week over a period of three months. Patients were encouraged to repeat the exercises at home on a daily basis. After 12 weeks, patients were instructed perform home exercises two to three times per week.

Arthroscopic subacromial decompression (ASD) and rehabilitation

In all patients in the surgery treatment group, an investigation of the stability of the shoulder joint under general anaesthesia was performed followed by arthroscopic examination of the gleno-humeral joint, the rotator cuff, and the subacromial bursa. All three surgeons (Peter Biberthaler, Volker Braunstein, Ernst Wiedemann) performing the procedures are experienced shoulder surgeons. The ASD was performed with the scope positioned in the posterior portal and included a bursectomy and a release of the coraco-acromial ligament. Acromioplasty was performed, transforming the undersurface of the acromion into a flat type I acromion. Spurs or osteophytes on the undersurface of the AC joint were debrided. Patients started physiotherapy on the day after surgery and performed a standard rehabilitation protocol starting with active assisted range of motion on day 1. With decreasing pain, this training was progressed with strengthening exercises of the rotator cuff and shoulder muscles.

Outcome measurements

The Munich Shoulder Questionnaire (MSQ) presents an innovative tool for self-assessment of shoulder function. It was especially designed for an effective follow-up of shoulder patients allowing for a quantitative assessment of the Constant, Shoulder Pain and Disability Index (SPADI) and Disabilities of the Arm, Shoulder and Hand (DASH) score consisting of a 30-items questionnaire. The MSQ had been validated previously and its accuracy and effectiveness for follow-up was demonstrated [20].

Statistics

Data are given as median values (25–75 % interquartile ranges). The results of the different groups were compared for each indicator using the Mann–Whitney U test. The level of significance was set at p < 0.05. Statistical analysis was performed using Sigma Stat 3.5 software (Systat® Inc., Chicago, IL, USA).

Results

Patient collective

Between January 2000 and December 2009, a total of 1.257 patients were identified. After exclusion, 331 with isolated SIS remained (26.3 %). Follow-up data for 307 (92.8 %) (151 males, 156 females) of these patients were recorded. The 165 (53.8 %) patients were treated conservatively and 142 (46.2 %) surgically by ASD. The median age was 57 (25 %–75 %: 48–63) years. The median interval between initiation of therapy and follow-up was 55 (25 %–75 %: 25–87) months. Group I (age < 57 yearrs.) consisted of 165 patients with a median age of 49 (25 %–75 %: 42–53) yearEighty-two(49.7 %) of the group I patients were treated conservatively and 83 (50.3 %) received ASD, respectively. In contrast, group II (age > 57 years) comprised 142 patients. Their median age accounted for 63 years and (25 %–75 %: 60–67). 83 (58.5 %) patients underwent conservative therapy, whereas 59 (41.5 %) underwent surgery by ASD (for patient characteristics see Table 1).

Table 1 Outcome
Full size table

Outcome measures—Overall data

The median MSQ score in patients (> 57 year.) of group I amounted to 83 points (25 %–75 %: 65–92), in group II (age > 57 year.) the MSQ score accounted for 84 points. (25 %–75 %: 67–93). Equivalent results were found for the DASH (group I: 12 pts.; group II: 10 pts.), the SPADI (group I: 88 pts.; group II: 91 pts.) and the Constant score (group I: 74 pts.; group II: 74 pts.) (overall data not shown).

Outcome measures—Conservative vs. ASD treatment

The subgroup analysis of group I (age < 57 years.) revealed a MSQ of 84 points (25 %–75 %: 65–93 pts.) in conservatively treated patients and of 83 points (25 %–75 %: 66–91 pts.) in patients after ASD (p = 0.37). Concurrently, no difference was found for the DASH score (conservative: 12 pts. 25 %–75 %: 1–32 pts.; ASD: 12 pts. 25 %–75 %: 5–37 pts.) as well as for the SPADI score (conservative: 88 pts. 25 %–75 %: 66–98 pts.; ASD: 87 pts. 25 %–75 %: 64–96 pts.). There was also no significant difference in the Constant score comparing conservatively and surgically treated patients (75 pts. 25 %–75 %: 64–83 pts. vs. 73 pts. 25 %–75 %: 61–80 pts.) (see Fig. 1).

Fig. 1
figure 1

Outcome results of patients younger than 57 years (group I). Light grey: conservative treatment; dark grey: Arthroscopic subacromial decompression (ASD) Data are given as vertical boxplots (median: horizontal boxline; 25–75 % interquartile ranges; standard deviations: horizontal line). There is no statistical significant difference between the sub-collectives. Abbreviation: MSQ Munich Shoulder Questionnaire; DASH: Disability of the Arm, Shoulder and Hand; SPADI: Shoulder Pain and Disability Index; CS: Constant Score

Full size image

In contrast, the analysis of group II (age > 57 years.) showed a significantly lower MSQ score of 81 points (25 %–75 %: 64–90 pts.) in patients who had had physiotherapy, in comparison to those who had undergone ASD with 89 points (25 %–75 %: 75–94 pts.) (p < 0.05). This significant difference was also found when comparing all other evaluated outcome parameters (DASH: 15 pts. 25 %–75 %: 4–29 pts vs. 5 pts. 25 %–75 %: 1–22 pts; SPADI: 87 pts. 25 %–75 %: 65–97 pts vs. 95 pts. 25 %–75 %: 77–100 pts; Constant: 71 pts. 25 %–75 %: 59–78 pts vs. 77 pts. 25 %–75 %: 63–83 pts; for data see Fig. 2).

Fig. 2
figure 2

Outcome results of patients older than 57 years (group II). Light grey: conservative treatment; dark grey: Arthroscopic subacromial decompression (ASD) Data are given as vertical boxplots (median: horizontal boxline; 25–75 % interquartile ranges; standard deviations: horizontal line). * p < 0.05 group conservative vs. group ASD, Mann–Whitney U. Abbreviation: MSQ Munich Shoulder Questionnaire; DASH: Disability of the Arm, Shoulder and Hand; SPADI: Shoulder Pain and Disability Index; CS: Constant Score

Full size image

Discussion

In this study, we demonstrated a retrospective long-term follow-up of patients having been treated for SIS either conservatively by physiotherapy or arthroscopically by ASD. Regarding the impact of age on the functional outcome, we found superior results in patients older than 57 years of age following ASD in comparison to conservative treatment (p < 0.05). In patients younger than 57 years, equivalently good to excellent results were found for both treatment regimens.

Subacromial impingement syndrome

Different theories regarding the pathogenesis of SIS and rotator cuff tear (RCT) have been suggested in the last decades. According to Neer, SIS is caused by extrinsic compression and mechanical wear of the RC against the coraco-acromial arch during elevation. This subject of extrinsic impingement was supported by Bigliani et al., who added their results on the correlation of the prevalence of RCT and the acromion shape. In this context, the propagation of ASD by eliminating the extrinsic factors is supported by a considerable number of clinical studies. However, intrinsic degenerative processes within the hypovascular zone on the greater tuberosity obviously contribute to RCT as well. In this context, Ozaki et al., reported that although a lesion in the anterior third of the undersurface of the acromion was always associated with a tear of the rotator cuff, the converse was not true [10]. Summarising, the pathogenesis of SIS is not yet fully understood, but most authors describe it as a combined result of extrinsic and intrinsic mechanisms, contributing to RCT [8–11].

Patients

In this study patients with an age between 20 and 82 years were included. The lower age limit is in accordance with previous studies, although it is acknowledged that glenohumeral instability is the most common cause of symptoms in younger patients. Regarding the upper age limit, most authors did not report a clear cut-off. Therefore, in most studies, the mean age ranged from 42 to 57 years. Beyond doubt, the utmost problem of patients >70 years is not an isolated SIS, but degenerative RCT. In this context, the dropout of >70 % of patients after following the exclusion criteria was mainly caused by confounding RCTs. Though, 142 patients >57 years with isolated SIS have still been included.

Reviewing the further inclusion criteria, several authors emphasised the impact of duration of symptoms on the outcome [21, 22]. In this context, Patel et al., have analysed some of the factors possibly influencing the results of ASD and pointed out that besides the response to the impingement test, especially the duration of symptoms of over one year is correlated to a significantly worse outcome. Further criteria such as occupation or insurance status, that were not considered in our study, have been previously shown to be of inferior relevance [16].

Conservative treatment

Patients having been treated conservatively received a standard physiotherapy for 12 weeks on an outpatient basis. The protocol with focus on strengthening of the rotator cuff and the scapular stabilising muscles is comparable to previous studies. Although patients thereby did not undergo formal training by physiotherapists, we regard this to be not statistically significant. In this context, Andersen et al., in 1999 analysed the effectiveness and outcome following a physiotherapeutically supervised vs. an unsupervised self-training following ASD and demonstrated clearly significant poorer results in the supervised group [23].

Arthroscopic subacromial decompression

The ASD, as performed in our study, is an established technique, having been used by the surgeons involved for several years. This is of distinct importance as several authors clearly proved that there is a learning curve associated with the technique of ASD and underlined this with inferior results in their own patients and a significant correlation of the learning curve and improving clinical results. Concerning the question of lateral clavicle resection, patients with AC joint arthritis and consecutive AC joint resection have been excluded in our study. This is in line with the results of Kharrazi et al., who do not recommend routine AC joint resection at the time of subacromial decompression but favoured a removal of the osseous excrescences on the undersurface of the joint.

Limitations

This study has distinct weaknesses. The first one is of course its retrospective nature. Facing the high number of enrolled patients and strict exclusion criteria, we nevertheless regard it as a significant contribution to the knowledge of the outcome of SIS. The major drawback is of course that we cannot exclude the possibility that bias in patient selection might have influenced the results of the older patients. As we found no difference in the young patients comparing both treatment regimens, we however regard the selection bias to be negligible.

Conclusion

Accurate diagnosis of the impingement syndrome requires a thorough patient history and a careful clinical examination followed by special investigations to exclude other conditions. According to our data, arthroscopic subacromial decompression in older patients with isolated subacromial impingement seems to be a promising therapeutic tool.