Introduction
Knee arthroplasty is one of the most common operations in Germany, with around 170000
procedures carried out in 2021 [1 ]. In Germany, 87% of procedures are total knee arthroplasties (TKA) and 13% are unicondylar
knee arthroplasties (UKA) [2 ].
There are several conservative and surgical treatment options to treat osteoarthritis
of the knee which can be used depending on the stage of disease [3 ]. In the early stages, the focus must be on educating the patient, lifestyle adaptations,
supervised exercises, and weight loss. It is also important to investigate whether
reconstructive procedures (e.g., osteotomy) could be an option. If the symptoms progress,
treatment can consist of medication (especially topical and systemic non-steroidal
anti-inflammatory drugs [NSAIDs] after all contraindications have been taken into
account) and non-medical therapies (e.g., physiotherapy, walking aids). A recently
published network meta-analysis which was based on a large number of cases (152 randomised
studies [RCT], n = 17431 patients with hip or knee osteoarthritis) showed that supervised
exercises achieved results which were equivalent in terms of pain relief and functional
improvement
within 6 months to those obtained with NSAIDs/paracetamol [4 ]. Based on these findings, supervised exercises should be given a high priority in
non-surgical therapy, particularly considering the potential side effects of taking
medication which can never be totally excluded.
For patients with moderate or severe osteoarthritis of the knee and symptoms that
cannot be relieved sufficiently with non-surgical therapy, knee arthroplasty is one
of the most successful and effective treatment options [5 ]
[6 ]
[7 ]. The patient-related results reported in the National Joint Registry (NJR) demonstrate
that 95% of patients experience an improvement with regards to knee pain and function
(by an average of 17 points on the Oxford Knee Score) [8 ]. According to a meta-analysis published in 2019, the expected durability of a TKA
is 93.0% after 15 years, 90.1% after 20 years and 82.3% after 25 years [9 ]. Moreover, the rate of surgery-related complications necessitating a longer stay
in hospital (1.26% for general complications and 0.97% for specific
complications) and the mortality risk (0.04%) are extremely low [10 ].
In Germany, the incidence of knee arthroplasty varies according to region and can
differ between individual federal states by a factor of up to 3.2, with higher numbers
of procedures reported for South Germany [11 ]. It has been suggested that the reason for this variation has been the lack of standardised
decision criteria which would ensure that the indications for knee arthroplasty are
transparent and consistent [11 ]. This prompted specialists to launch the guideline project “Evidence- and Consensus-based
Indication Criteria for Knee Arthroplasty (EKIT-Knee)” under the aegis of the German
Society for Orthopaedics and Trauma Surgery (DGOU) and the German Arthroplasty Society
(AE) [12 ], which led to a guideline being developed. This guideline has now been updated [7 ]. The aim of the update was to review
recommendations based on current evidence and bring them into line with the existing
S3-guideline “Evidence- and Consensus-based Indication Criteria for Total Hip Arthroplasty
for Coxarthrosis (EKIT-Hip)” [13 ].
Guideline recommendations
The following five main criteria must be met for indication of a knee arthroplasty:
structural damage, knee pain, insufficient conservative therapy, reduced quality of
life, and subjective psychological stress.
Structural damage
Structural damage is present if X-ray imaging shows Kellgren-Lawrence grade 3 (moderate)
or 4 (severe) osteoarthritis or if osteonecrosis has resulted in deformation or a
defect of the joint surface. A posterior-anterior (PA) weightbearing radiograph of
the knee (Rosenberg view) is recommended as this permits better assessment of the
loss of joint space in the posterior femoral subregions, especially in cases with
valgus osteoarthritis.
As this procedure is irreversible and because of the potential risks associated with
the procedure, surgery is usually only indicated for cases with moderate or severe
osteoarthritis. Two recent meta-analyses have shown that a significantly higher pre-/postoperative
improvement of pain, knee function and satisfaction is achieved in cases with moderate
or severe ostearthritis compared to mild osteoarthritis (n = 8542 knee arthroplasties)
[14 ] and that surgery for mild or moderate gonarthrosis was associated significantly
more often with chronic pain and dissatisfaction (n = 12723 knee arthroplasties) [15 ]. Nevertheless, there are some borderline cases where, despite radiological imaging
showing a less severe level of osteoarthritis, the full extent of cartilage damage
(full-thickness damage) is only visible on MRI and knee arthroplasty is indicated.
Additional imaging using MRI must only
be done if there are discrepancies between the clinical and the radiological findings.
This also applies to osteonecrosis, where knee arthroplasty may be indicated even
if there are no signs of advanced osteoarthritis. MRI is particularly useful to show
extensive full-thickness cartilage damage and the extent of osteonecrosis or to assess
whether bone marrow oedema could be present.
Knee pain
Many patients with advanced osteoarthritis report knee pain. Medication is usually
very effective to treat short-term pain. If the knee pain cannot be improved by guideline-appropriate
non-surgical therapy [16 ] administered over a period of at least 3 months, it can be asssumed that further
non-surgical therapy measures will also not be able to improve symptoms. Patients
may then be offered knee arthroplasty. Almost all patients report pain prior to knee
arthroplasty and pain is also an important criterion for the decision to have surgery
[17 ]. Almost all patients expect that knee arthroplasty will result in pain relief and
for most patients, this expectation is fulfilled or even exceeded [18 ].
Inadequate guideline-appropriate non-surgical therapy
The recommendation that non-surgical treatment of patients with osteoarthritis should
consist of a combination of medication-based and non-medication-based therapeutic
measures is based on a broad international consensus [6 ]
[19 ]
[20 ]
[21 ]. Patients must have received or been offered the following non-medication-based
measures:
Patient education (information, education, and counselling about the disease)
Exercise therapy and promotion of physical activity
Weight loss for patients who are overweight or obese
The evidence base for the 2 first two measures is extensive and of high quality. Disease-specific
information (therapy programmes, self-management programmes, tutorials, booklets)
significantly affects pain relief, reduces pain medication, and improves quality of
life and self-help [20 ]. The guideline of the Osteoarthritis Research Society International (OARSI) reported
on 8 RCTs which showed that exercise therapy had a significant impact on pain relief,
functional improvement, and activity levels [6 ].
If there is insufficient improvement despite carrying out guideline-appropriate non-surgical
therapeutic measures for at least 3 months and a high subjective level of suffering
persists, knee arthroplasty is indicated. An analysis of 77 RCTs (comparison of an
intervention group which received additional exercise therapy with a control group
which did not receive additional therapy) by Goh et al. [22 ] demonstrated that the positive effects with regards to pain relief, improved function
and activity, and improved quality of life following additional exercise therapy peaked
after 8 weeks and began to decrease thereafter. After about 9–18 months no differences
could be found between the intervention group and the control group. It can therefore
be assumed that if guideline-appropriate conservative therapy does not result in sufficient
improvement within 3 months, an improvement will not occur later on either.
Reduction in health-related quality of life and subjective level of suffering
After assessing the severity of osteoarthritis both clinically and radiologically,
the adverse effects and the level of suffering caused by the osteoarthritis must also
be evaluated. In addition to pain, this includes functional limitations, adverse effects
on activities of daily living and a reduced health-related quality of life [19 ]
[23 ]
[24 ].
The collection of this data is not just relevant when evaluating treatment outcomes
but also important when making shared decisions for or against knee arthroplasty.
It is important to do this by using validated instruments for patient-reported outcome
measures (PROMs). The AE has published consensus-based recommendations on measuring
outcomes after hip and knee arthroplasty procedures [24 ]. Although the recommendation to use the Oxford Knee Score (OKS) or alternative instruments
to measure outcomes (WOMAC, KOOS or KOOS-PS) along with the use of a generic score
(e.g., EQ-5D, SF-12, SF-36) was primarily aimed at clinical studies, the recommendation
can also be expanded to cover the use of such scores in general practice. The above-listed
PROMs are also available in German [24 ].
Ultimately, the indication for knee arthroplasty must be based on a high level of
suffering with knee-related complaints (pain, functional limitations, restrictions
with regards to activities of daily living) and a reduced health-related quality of
life combined with the presence of other indication criteria (Kellgren-Lawrence grade
3 or 4 osteoarthritis, guideline-appropriate conservative therapy for at least 3 months
without sufficient improvement) [7 ]
[20 ].
If not all of the main criteria are met, subsidiary criteria can also be used to confirm
the indication for knee arthroplasty. Pronounced misalignment or instability due to
destruction of the joint despite limited pain can be an indication that joint replacement
is required. This must be reviewed on a case-by-case basis and the findings recorded.
Contraindications
The number of patients for whom knee arthroplasty is absolutely contraindicated is
relatively small. Because of the increased risk of infection, revision, and mortality,
knee arthroplasty is not indicated or should be critically reviewed if the following
factors are present:
Absolute contraindications:
Active, not fully healed infection
Acute or chronic comorbidities which constitute a contraindication for elective surgery
(e.g. acute cardiovascular event)
Relative contraindications:
Significantly reduced life expectancy due to comorbidities (in such cases, the benefits
and risks of surgery must be weighed up particularly carefully)
Morbid obesity (BMI ≥ 40 kg/m2 )
While most contraindications are undisputed, the inclusion of morbid obesity is regularly
discussed. The association between BMI ≥ 40 kg/m2 and a significantly increased risk of periprosthetic infections is not in doubt [25 ]
[26 ]. The association between higher BMI and arthroplasty failure has also been clearly
proven in the German Arthroplasty Registry (EPRD) [2 ]. However, when patients have a BMI ≥ 40 kg/m2 , then achieving a relevant weight loss is very difficult. To what extent weight loss,
for example achieved with the help of bariatric surgery, ultimately reduces the risk
of knee arthroplasty is currently not known. It is therefore important to be especially
critical when weighing up the benefits against the risks of arthroplasty in these
patients. The recommendation is that in all such cases, the patient should see an
obesity outpatient clinic. It is also important to be aware that with these patients,
implantation of an arthroplasty could constitute an off-label use, i.e., the implant
is being using outside its approved use, as the instructions for use (IFU) for different
knee arthroplasty systems cite obesity or morbid obesity as contraindications. If
necessary, patients must be informed about this off-label use [27 ]. As the functional gains and the satisfaction with the arthroplasty are comparable
for obese and normal-weight patients, it seems to be ethically indefensible to generally
exclude these patients from knee arthroplasty despite the higher risks involved [28 ]
[29 ]
[30 ].
Optimisation of modifiable risk factors
Patients have individual risk factors which can negatively affect the perioperative
and postoperative complication rates as well as treatment outcomes and the revision
rates [2 ]
[31 ]. When planning surgery, it is important to know whether existing risk factors can
be modified and individual risk of complications reduced. Modifiable risk factors
include uncontrolled diabetes mellitus, obesity, smoking, anaemia, intraarticular
injection of glucocorticoids, mental disorders, and active inflammatory rheumatic
disease.
When treating patients with diabetes mellitus, blood glucose levels must be optimally
controlled prior to knee arthroplasty. The aim should be to achieve an HbA1c of less than 8%.
If the patient has a BMI of ≥ 30 kg/m2 , the patient should be advised to lose weight.
Smokers must be advised to stop smoking at least 1 month prior to the planned surgical
procedure.
A diagnostic workup to check for anaemia should be done prior to knee arthroplasty
and treatment should be initiated for patients with anaemia requiring treatment.
Following intraarticular injection of glucocorticoids, surgery can be carried out
after 6 weeks at the earliest; the recommended interval between injection and surgery
should be 3 months.
If there is a suspicion that the patient has a mental disorder which has not been
adequately treated, the patient should be advised to have a consultation with a mental
health specialist prior to knee arthroplasty.
Active inflammatory rheumatic disease must be adequately controlled using medication
prior to knee arthroplasty. If glucocorticoids are required at the time of surgery,
the target must be a glucocorticoid dose of not more than 7.5 mg prednisolone per
day.
The evidence for the respective recommendations is presented in the guideline [7 ].
Additional risk factors
There are additional risk factors which are associated with a higher complications
profile or poorer patient-relevant outcomes. Many of these risk factors cannot be
optimised. The following risk factors must be taken into consideration and discussed
with the patient when making the indication foor knee arthroplasty:
Prior infection of the knee joint
Increased risk of infection
Higher perioperative risk (ASA 3 and 4)
Other physical or psychological comorbidities or medications which increase the surgical
risk
Substance addiction or abuse (including nicotine, alcohol)
Unrealistic expectations from the patient
Shared decision-making
Shared decision-making improves the outcome after knee arthroplasty [32 ], but it is not yet part of general clinical practice [33 ]. Fulfilling patients’ expectations on knee arthroplasty is an essential factor for
patient satisfaction [18 ]
[34 ]. For this reason, identifying and recording the individual patient’s expectations
and goals is an important part of shared decision-making as is the subsequent discussion
about whether knee arthroplasty can meet the patient’s expectations. The discussion
should include expected effects with regards to postoperative outcome (pain relief,
improved functionality, activity, and quality of life), general surgical risks, the
patient’s individual risk profile, and the probability that the individual goals will
be fulfilled. Information materials which can be easily
understood by patients should be available to support the information given to the
patient. Ultimately, the consultation will lead to a shared decision for or against
surgery. There should be agreement that the expected benefits of surgery will outweigh
the possible risks.
The contents of the S2k-guideline on indications for knee endoprosthesis have been
summarised in a checklist to allow the recommendations to be easily transferred into
routine clinical and surgical practice ([Fig. 1 ]).
Fig. 1 Checklist of indications for knee arthroplasty.
