Background
Arthrofibrosis (AF) is one of the most frequent complications after knee surgery, such as anterior cruciate ligament repair or total knee arthroplasty (TKA). Concerning the epidemiology, the reported incidences range from 1–35% [1‐6]; however, several authors also included patients with temporary stiffness after TKA. Therefore, the real incidence seems to be 3–4% [3].
Tightness as well as a painful limitation of joint mobility already develop in the early postoperative phase due to an increase in fibrous tissue [7].
Anzeige
Function is already impaired at early stages of AF: even in patients with a slight decrease of extension walking is affected and reduced flexion leads to difficulties in climbing stairs or sitting down. These limitations lead to impairments in instrumental activities of daily life, such as driving a car as well as in participation, including returning to work, with overall low quality of life [8].
Currently, the treatment is based on a pathophysiological adhesion model and usually involves forced stretching to regain the range of motion; however, intense classical physiotherapy including stretching or continuous passive motion is often painful and does not improve the range of motion but frequently even leads to a further decrease in function. In addition, manipulation under anesthesia and even surgical arthrolysis often only produce a short-term improvement in mobility and lead to unsuccessful treatment processes [9].
Relatively recent research advocates a new understanding of the so-called cellular cytokine-based pathogenesis of AF, which involves several proinflammatory cytokines, myofibroblasts and emotional distress [8, 9]. Therefore, several important aspects need to be acknowledged in the rehabilitation process, such as the warning function of pain, termination of forced joint mobilization, permitted motion within a pain-free range, balancing of the vegetative system and relaxation techniques [9, 10]. Currently, there is only little work in the literature on comprehensive conservative concepts for the treatment of AF patients.
Based on findings by Kieslich and Rückert [10] as well as on those by Traut [9], our clinic established a comprehensive antifibrotic treatment protocol to meet the special needs of patients with knee AF. The aim of this retrospective analysis was to evaluate the patients treated up to now.
Anzeige
Methods
This retrospective study was conducted at the Clinic Donaustadt, an academic teaching hospital of the Medical University of Vienna, Austria. Before reviewing the records of the patients treated according to the clinic’s AF protocol, the approval by the local ethics committee was obtained. This research was performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments.
Before diagnosing AF, other causes of knee stiffness had been ruled out, such as periprosthetic joint infection, implant instability, malpositioned or wrongly sized implants, neuroma, retropatellar arthrosis, complex regional pain syndrome and somatoform disorders. As there is no consensus on the clinical diagnosis of AF, the criteria were defined based on the recent literature [3, 9]. First, difficulty of pain control, particularly during mobilization; second, prolonged reduction in range of motion; third, feeling of tightness and fourth, an immobile patella.
Based on previous observations by Kieslich and Rückert [10], as well as by Traut [9] the protocol includes:
-
patient education,
-
cessation of painful stretching,
-
soft gravity-based relaxation/stretching,
-
passive patella mobilization,
-
improvement of microcirculation by manual lymphatic drainages/connective tissue massages/painless electric stimulation (e.g., high tone therapy osteoarthritis program)/laser therapy/magnetic field therapy/reflexology/osteopathic soft tissue techniques,
-
general relaxation techniques, autogenic training (with the aim of vagal nerve stimulation and sympathetic inhibition),
-
medical training therapy (e.g., ergometer, resistance training), subaqua therapy, continuous passive motion (CPM; 30 min twice daily over 6 weeks at low velocity), gait training, mirror therapy (in cases of concomitant complex regional pain syndrome) strictly in a pain-free range and with no increase of feeling of tightness,
-
no additional surgical procedure for at least 1 year.
The following parameters were investigated at baseline, after 2 months of treatment as well as at a follow-up examination after at least 12 months:
-
patellar mobility (mediolateral and craniocaudal),
-
range of motion,
-
feeling of tightness,
-
overheating,
-
nocturnal pain, pain after activity, pain medication,
-
general mobility, assistive devices such as crutches,
-
working status.
-
necessity for additional surgery.
In addition, overall satisfaction with the program was obtained via a 5-tier scale (1 = very unsatisfied, 5 = very satisfied).
Statistics
Due to the small number of patients, data are presented descriptively. Range of motion was assessed via the neutral zero method, feeling of tightness via yes/no, clinical difference in skin temperature compared to the contralateral limb via yes/no. Mobility, necessity of aids and working status were described nominally. The necessity for operative revision was assessed via yes/no.
Results
To date, four patients have been treated according to the AF protocol.
The baseline data are provided in Table 1.
Table 1
General medical data is presented for the four patients, as well as detailed information on baseline mobility and clinical symptoms
Patient 1 | Patient 2 | Patient 3 | Patient 4 | |
|---|---|---|---|---|
Gender, age (years) | M, 68 | F, 50 | M, 53 | F, 59 |
Cigarette smoking | – | – | 20/day | – |
Previous surgery | – | Anterior cruciate ligament repair (15 years ago) | Arthroscopic partial meniscectomy (3 years ago) | Mosaicplasty (14 years ago) |
Chronic conditions | Atrial fibrillation | – | – | – |
Surgery | Total knee arthroplasty; manipulation under anesthesia (6 weeks after surgery) | Total knee arthroplasty | Arthroscopic partial meniscectomy | Total knee arthroplasty; manipulation under anesthesia (8 weeks after surgery) |
Time since surgery | 2 months | 1 month | 2 months | 3 months |
Patella mobility medial-lateral (cm) | 0.5 | 0.5 | 0 | 0 |
Patella mobility craniocaudal (cm) | 0 | 0 | 0 | 0 |
Range of motion (degrees; neutral zero method) | 0‑5-80 | 0‑15-85 | 0‑20-90 | 0‑15-90 |
Tightness | Yes | Yes | Yes | Yes |
Overheating | Yes | Yes | Yes | Yes |
Night pain | Yes | Yes | Yes | Yes |
Pain after activity | Yes | Yes | Yes | Yes |
Analgesics | No | Yes (metamizole, tramadol) | Yes (mefenamic acid) | Yes (naproxen, metamizole, tramadol) |
Aids | Yes (crutches) | Yes (crutches) | Yes (crutches) | Yes (crutches) |
Working status | Active | Sick leave | Sick leave | Sick leave |
Anzeige
In two patients manipulation under anesthesia had already been performed, however, unsuccessfully. In patient 1 the range of motion before manipulation under anesthesia was 0‑10-65° and in patient 4 it was 0‑20-70°.
All patients had highly impaired craniocaudal patellar mobility and knee extension was incomplete. Moreover, all patients had a feeling of tightness, overheating compared to the contralateral knee, night pain as well as pain after activity.
After 2 months of treatment according to the AF concept, patella mobility increased in all patients: Mediolateral mobility doubled in two and tripled in one patient and craniocaudal mobility increased from 0 to 0.5 cm in three out of four patients.
The range of motion increased in most patients (15–35°), except for patient 4 where adherence to the program was low and she reported painful vigorous stretching exercises on a regular basis.
Anzeige
One patient reported the disappearance of tightness and overheating and two patients had no nocturnal pain anymore. One patient was pain-free after activity.
All patients discontinued both analgesics and walking aids.
One patient returned to work (Table 2).
Table 2
Clinical course after 2 months
Patient 1 | Patient 2 | Patient 3 | Patient 4 | |
|---|---|---|---|---|
Patella mobility medial-lateral (cm) | 1 | 1.5 | 1 | 0.5 |
Patella mobility craniocaudal (cm) | 0.5 | 0.5 | 0.5 | 0 |
Range of motion (neutral zero method) | 0‑0-110 | 0‑10-95 | 0‑15-100 | 0‑15-90 |
Improvement in extension (°) | 5 | 5 | 5 | 0 |
Improvement in flexion (°) | 30 | 10 | 10 | 0 |
Tightness | Yes | Yes | No | Yes |
Overheating | Yes | Yes | No | Yes |
Night pain | Yes | No | No | Yes |
Pain after activity | No | Yes | Yes | Yes |
Analgesics | No | No | No | No |
Aids | No | No | No | No |
Working status | Active | Sick | Sick | Sick |
At the follow-up examination (Table 3) patella mobility further increased: Mediolateral mobility was 1.5–2.0 cm and craniocaudal 1.0–2.0 cm.
Table 3
Follow-up after 12 months
Patient 1 | Patient 2 | Patient 3 | Patient 4 | |
|---|---|---|---|---|
Patella mobility medial-lateral (cm) | 1.5 | 2.0 | 1.5 | 1.5 |
Patella mobility cranio-caudal (cm) | 1.0 | 2.0 | 1.0 | 1.0 |
Range of motion (neutral zero method) | 0‑0-120 | 0‑0-125 | 0‑0-140 | 0‑15-100 |
Improvement in extension (°) compared to baseline | 5 | 15 | 20 | 0 |
Improvement in flexion (°) compared to baseline | 40 | 40 | 50 | 10 |
Tightness | No | No | No | Yes |
Overheating | No | No | No | Yes |
Night pain | No | No | No | Yes |
Pain after activity | No | No | No | No |
Analgesics | No | No | No | No |
Aids | No | No | No | No |
Working status | Active | Active | Active | Sick |
Additional surgery | No | No | No | Revision surgery (at another clinic) |
Anzeige
Knee extension was complete in three out of four patients, and in most patients, flexion increased to at least 120°. In most patients, tightness, overheating, nocturnal pain and pain after activity had disappeared. None of the patients had analgesics or walking aids and three out of four patients had returned to work. Functionally, three out of four patients rated no limitations in daily life, including recreational activities. In particular, patient 1 returned to dancing on a regular basis. Patient 2 returned to recreational cycling and swimming. Patient 3 restarted long hiking tours.
In overall ratings, three out of four patients expressed high satisfaction (5 of 5 points) with the AF program. Patient 4 rated 3 of 5 points, i.e., neutral.
Concerning additional surgery, only patient 4 had an additional operation, i.e., revision surgery at another clinic. According to the other clinic’s discharge letter, the prosthesis had been too large and a CT-adjusted prosthesis was implanted. Nevertheless, neither range of motion nor the symptoms had clearly improved. The adherence to the program was still low and she continued painful vigorous stretching exercises on a regular basis.
Discussion
This pilot evaluation demonstrated a high potential of the conservative AF concept. A clear improvement in most patients was already observed after 2 months, with further enhancement at the follow-up examination. Out of the four patients three were overall highly satisfied with the AF program and could resume recreational activities such as long hiking tours and dancing.
Patient 4 reported not only low adherence to the AF concept but also had a revision surgery at another clinic; however, thereafter neither the symptoms nor range of motion clearly improved. Therefore, we hypothesize that a significant part of her condition could be alleviated by adhering to our clinic’s AF protocol.
In order to facilitate a good prognosis it is essential to identify AF as early as possible. Even though the clinical diagnostic criteria of AF are not uniform in the literature some parameters are frequently mentioned [1, 3, 9]. First, pain control is difficult, particularly during mobilization. According to our observations there were high pain intensities even within the first postoperative week. Second, knee range of motion is decreased with a stronger decrease referring to more severe fibrosis. Third, the feeling of tightness is typical for AF. Fourth, an immobile patella is characteristic for AF.
According to an international consensus classification based on the extent of movement limitation, our patients had moderate AF [1]. In addition, potential biomarkers for AF after total knee arthroplasty, such as transforming growth factor-beta receptor 1, were immunohistochemically identified [5].
Before diagnosing AF, other causes of knee stiffness need to be ruled out, such as periprosthetic joint infection, implant instability, malpositioned or wrongly sized implants, neuroma, retropatellar arthrosis, complex regional pain syndrome, or somatoform disorders [3, 9].
Several invasive treatment options were reported, such as manipulation under anesthesia, arthroscopic or open adhesion lysis, as well as revision total knee arthroplasty [3]. Some orthopedic surgeons may use dexamethasone and ketorolac perioperatively when performing manipulation under anesthesia according to the Mayo Clinic protocol [2].
In two of our patients, manipulation under anesthesia had already been performed (without dexamethasone or ketorolac). Some patients may benefit from manipulation under anesthesia. According to a recent review [3], the average gains in range of motion are 22–39° for flexion and 2–5° for extension, with failure rates of 7–24%. Even though on a low level potential complications, such as fractures, wound dehiscence, patellar tendon avulsion, and hemarthrosis need to be considered.
In addition, other authors reported a high relapse rate after manipulation under anesthesia [9] and in both of our patients, the procedure was unsuccessful.
In contrast, the conservative AF program was effective: In most patients, clear gains of 10–30° for flexion and 5° for extension were already observed after 2 months, with further increases to 40–50° for flexion and 5–20° for extension after 12 months.
Therefore, the cellular cytokine-based model of pathophysiology needs to be kept in mind. According to this model, mechanical stress (such as vigorous stretching exercises or manipulation under anesthesia) can maintain the activation of myofibroblasts and thereby AF. This model can explain the unsuccessful course in some patients after manipulation under anesthesia and in these patients our clinic’s concept may be a reasonable alternative.
There is relatively little work on comprehensive conservative treatment concepts [9, 10]. Some authors advocate aggressive physiotherapy [3]; however, aggressive approaches may again trigger myofibroblasts due to high mechanical stress. The debate on CPM is controversial [3]. In our patients, CPM was performed for 30 min twice daily over 6 weeks carefully and at low velocity without provoking pain or tightness.
Pharmacological interventions, such as prednisolone and propranolol are off-label symptomatic treatments without directly targeting myofibroblasts. Therefore, they are not classified as causal treatment options and their effect seems to be lower compared to correct behavior strategies and physiotherapy [9]. Therefore, drugs were not included in this AF concept at our clinic.
Based on previous observations [9, 10], we established a comprehensive AF program at our clinic. According to this protocol the first important step is patient education on AF causes and treatment options. This is the basis for achieving correct behavior strategies, such as the cessation of painful stretching. Moreover, soft gravity-based relaxation/stretching techniques are implemented as well as passive patella mobilization. These techniques must not cause any pain, otherwise AF could be maintained.
In addition, microcirculation should be improved, e.g., by manual lymphatic drainages, laser therapy, magnetic field therapy, osteopathic soft tissue techniques or painless electric stimulation (e.g., via the osteoarthritis program of high tone therapy). High tone therapy, a specific type of electrotherapy, delivers medium frequency alternating current. The carrier frequency ranges from 4 kHz to 33 kHz and both frequency and amplitude are modulated simultaneously (simulFAM ®; HiToP ® 4 touch, GBO Medizintechnik AG, Rimberg, Germany). In the so-called “osteoarthritis program”, both low and medium frequency currents are combined, thereby using both the low frequency gate control concept of analgesia, as well as the regenerative effects of the simulFAM i ® medium frequency current, as special form of simulFAM ® [11].
Furthermore, the vegetative nervous system should be balanced, e.g., by general relaxation techniques, reflexology, connective tissue massages or autogenic training (with the aim of vagal nerve stimulation and sympathetic inhibition).
Finally, medical training therapy is an important part of the program. Ergometer training and resistance training should be performed as well as subaqua therapy and gait training. Continuous passive motion (CPM) was part of the program as well and in cases of concomitant complex regional pain syndrome, also mirror therapy should be implemented. Again, it is emphasized that the training needs to be strictly kept in a pain-free range and the feeling of tightness must not increase. Finally, there should be no additional surgical procedure for at least 1 year.
In summary, the findings of this pilot evaluation are promising. Most patients were highly satisfied and clear improvements of the symptoms as well range of motion were observed. Therefore, this comprehensive AF program could serve as an alternative for those patients who fail to benefit from manipulation under anesthesia. Moreover, our concept could help to prevent the development of AF when implemented early in the course of the disease.
Therefore, our clinic’s concept should be evaluated in larger studies to confirm the positive effects. In particular, differences between patients with and without previous manipulation under anesthesia and with and without risk factors such as cigarette smoking and diabetes should be addressed as well as dose-response relationships of the different treatments, such as electrotherapy and continuous passive motion. Additionally, potential additional benefits by specific medications such as prednisolone and propranolol should be investigated.
Conflict of interest
R. Wakolbinger-Habel, J. Gaudernak, B.E. Scheffold, R. Fiala, R. Breuer, M. Bittner-Frank, C. Lang, H. Zehetner-Nics, A. Oljaca, M. Mousavi and T. Paternostro-Sluga declare that they have no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.