1.1

Introduction

Osteoarthritis represents the most frequent reason for performing total knee arthroplasty (TKA). At the outset, invalidating pain is treated with drugs and physiotherapy but the latter do not halt destruction of the knee joint. Hence, TKA is prescribed when the osteoarthritis is no longer controlled by medication or physical techniques and when functional activities (such as walking) are severely impaired . However, TKA can be a high-risk operation which may justify hospitalization of the subject in a rehabilitation service in order to achieve the best possible functional outcome . The TKA outcome is usually good to excellent, with less pain, improved functional abilities and better quality of life . Total knee arthroplasty patients show good longevity; the survival rate after 10 years is over 90% .

After TKA, rehabilitation care begins in the orthopaedic surgery service and continues (in most cases) with hospitalisation in an inpatient rehabilitation service or (less frequently) in a day hospital or even directly at home. Recently, Oldmeadow et al. developed and validated (in Australia) a method for predicting extended inpatient rehabilitation after hip and knee arthroplasty . The Risk Assessment and Predictor Tool (RATP) identifies three risk levels, with a correct discharge prediction rate of 74.6%. When the RATP score is below 6, hospitalization is predicted; when the score is over 9, home discharge is predicted. However, this score is not used in many other countries, where operated TKA patients are sent to a rehabilitation centre (although there are some exceptions). In fact, the discharge destination for TKA patients depends on several factors, such as age, comorbidities, postoperative functional ability, the availability of home help, the surgeon’s habits and the patient’s wishes . In France, a patient’s wishes are taken into account but do not prevail over the medical team’s discharge decision. Only a few studies have determined which factors contribute to rehabilitation hospitalisation and have focused on the length of stay; however, these were primarily health economic evaluations and did not consider the functional results on discharge from the rehabilitation centre . The objective of the present study was to collate the preoperative factors which may determine the duration of inpatient rehabilitation by taking into account the results of rehabilitation care (defined as 90 degrees of active knee flexion and a functional status allowing direct home discharge).

1.2

Method

1.3

Results

After exclusion of 38 patients having presented complications, a cohort of 244 TKA patients was described in terms of demographic parameters, preoperative functional ability and the presence or absence of a joint prosthesis or comorbidities ( Tables 1 and 2 ). On average, patients were admitted to the rehabilitation service 7.5 ± 1.7 days after surgery.

The duration of inpatient rehabilitation was 24.1 ± 8.1 [9–59] days and is presented as a histogram in Fig. 1 . The length of stay was similar in 2004 and 2007 ( F : 0.43; p : 0.72). On discharge from the rehabilitation service, active knee flexion was 97 ± 12 degrees, the walking distance was 369 ± 185 metres and the IKS function score was 48 ± 16 points.

Duration of inpatient rehabilitation (in days) with a Gaussian distribution of values.

After the univariate analysis, a linear relationship was found for only four preoperative factors: female gender, living alone, the presence of a home help and previous lower limb arthroplasty ( p < 0.25) ( Table 3 ). There was a significant interaction between living alone and female gender ( r : 0.309, p < 0.01) and between living alone and the presence of a home help ( r : −0.213; p < 0.05). This redundancy was confirmed in a colinearity analysis (tolerance > 3) and so the “living alone” parameter was excluded from the predictive model. We did not observe an interaction between female gender and the presence of a home help ( r : 0.024; p > 0.05). The “previous lower-limb arthroplasty” parameter was independent of the others ( Table 4 ). Neither patient’s age nor the length of stay in the orthopaedic service influenced the length of stay in the rehabilitation service. There were no relationships between the IKS function score, the Lequesne index and the length of stay in the rehabilitation service ( F : 0.02; p : 0.87 and F : 0.01; p : 0.90, respectively).

The multivariate linear regression analysis had low predictive power and included only female gender, the presence of a home help and previous lower limb arthroplasty ( r : 0.181; adjusted r ² : 0.02; standard error: 8.11) ( Table 5 ). The r ² value was 0.02, which means that only 2% of the variability in the duration of inpatient rehabilitation was accounted for by the predictive model. Hence, this model is not of use when observing a group of TKA patients and is not accurate enough to predict the outcome for a given patient ( p < 0.049).

1.4

Discussion

France’s High Authority for Health is currently promoting a decrease in the length of hospitalization and is encouraging home care. However, this reasoning must not decrease the quality of care and must take account of a patient’s social and economic status, as well as opportunities for outpatient rehabilitation . Our study sought to address the factors influencing the length of stay in a rehabilitation hospital, while considering the functional outcome of the rehabilitation care. The latter was defined as (a) the provision of a home help, (b) improvement in gait (with or without an assistive device) for minimizing the risk of falls and (c) active knee flexion of over 90 degrees on discharge. According to certain authors, the recovery of a stable medical state and functional ability is enough for home discharge ; however, we consider that active knee flexion of over 90 degrees is essential for home discharge, since some knee mobility is required after a TKA . The role of the kinesitherapy is thus to help the patients rapidly recover independent movement and transfer capacities and enough knee mobility to allow home discharge. Only Lin et al. have reported on the percentage of TKA patients able to bend their knee by more than 70 degrees . On discharge from the rehabilitation centre, only 74% of the study population had achieved this target, which corresponds to a rather poor result. In fact, this parameter is of great value because flexion of more than 90 degrees corresponds to a better functional result and enables greater independence (a more comfortable sitting position, easier transfers and use of stairs and the ability to step over an obstacle) . This factor accounts for the mean duration of inpatient rehabilitation of 24 days in our study, relative to the value of 11 days described by Lin et al. . However, the latter study focused on the length of stay in the rehabilitation hospital and did not taking into account the patients’ destinations on discharge – not all returned home directly, in contrast with our study population. Lin et al. explained the length of stay in the rehabilitation service in terms of age (over 80), male gender, black race, living alone, comorbidities and the presence of a TKA or total hip arthroplasty (THA) . However, the predictive model was very imprecise because it accounted for only 10% of the variability in the length of stay ( r ² : 0.10). Our study population was not the same because only Caucasian TKA patients were included. Furthermore, the two studies differed in terms of the post-TKA rehabilitation care; our objective was to recover good knee flexion and good functional status.

In our univariate analysis, four explanatory parameters (female gender, living alone, the presence of a home help and previous lower-limb arthroplasty) presented a significant linear relationship with the length of stay in the rehabilitation service. We were not surprised to note that the “living alone” parameter was specifically related to the need for a home help. The time needed during hospitalization to arrange a home help for those patients who required one after surgery could explain the inverse relationship and thus why the length of stay was an average of 1.5 days longer (representing an additional cost of 6%, i.e. 1.5 days relative to 24 days). This inverse relationship explains why the “living alone” factor was excluded from the predictive model.

We also identified an inverse relationship between previous lower-limb arthroplasty and the length of stay. By virtue reason of their prior experience of lower-limb arthroplasty, these patients may have benefited more from the rehabilitation programme, to the extent that their stay in the rehabilitation service was 1.5 days shorter, on average.

Various authors have shown a correlation between comorbidities (such as diabetes) and the length of stay required for rehabilitation . Our analysis may not have been precise enough for comorbidities to exert an influence. In our experience, daily treatment of diabetes (in the absence of other complications) does not result in an increase in the length of hospitalization. In contrast, we voluntarily excluded all acute complications from our study because the management of these patients could have increased the length of hospitalization. In these cases, the patients had to return to a surgical department or transfer to other medical services. In fact, our method distinguished between comorbidities with no impact on care management from those that do have an impact.

In the absence of post-TKA complications, the length of stay in the orthopaedic service did not influence the length of stay in the rehabilitation service. The preoperative walking distance, functional status and Lequesne index did not significantly influence the predictive model – despite the fact that a Lequesne index of over 12 is generally used to justify a TKA . The efficacy of the prosthetic surgery and the rehabilitation programme could explain this result because the patient’s state of health improved independently of the consequences of the osteoarthritis, as suggested by the rise in the IKS functional score from 35 to 48 points . However, this relatively low final score of 48 shows that our discharge criteria were not very severe in terms of knee flexion, walking ability and independence in the activities of daily living. Stricter criteria (taking into account knee extension, quadriceps strength and walking distance in the absence of gait aids or a helper) would perhaps have enabled better prediction. The patient’s pain level probably failed to have any effect, for the same reason. Hospitalization in a rehabilitation service can perhaps have a positive psychological effect on the patient, due to provision of support by the medical team.

1.5

Conclusion

Only 2% of the variability in the length of hospitalization for post-TKA rehabilitation is accounted for by female gender, the presence of a home help and previous lower-limb arthroplasty, when considering the results of the rehabilitation care and in the absence of complications. The predictive model is thus very limited and it was not possible to predict the length of post-TKA inpatient rehabilitation from the preoperative parameters studied here.

From an economic point of view, the potential for decreasing the length of stay in the rehabilitation service may depend specifically on the quality of the rehabilitation care in terms of improved knee flexion and the patient’s functional state. Arrangement of a home help led to a extra 1.5 days of hospitalization, i.e. a 6% increase in the cost of the hospital stay.

Appendix 1: the IKS function score

Appendix 2: the Lequesne index of severity for osteoarthritis of the knee

2.1

Introduction

La gonarthrose représente la maladie du genou la plus fréquente pour être à l’origine de la mise en place d’une prothèse totale de genou (PTG). Au début, les douleurs invalidantes sont traitées par des médicaments et des traitements physiques qui ne sont cependant pas capables de stopper la destruction de l’articulation du genou. La PTG est alors proposée quand la gonarthrose échappe au traitement médical et quand les activités fonctionnelles, telle la marche, sont altérées . La mise en place d’une PTG représente une chirurgie à risque qui peut justifier d’hospitaliser les patients en service de rééducation afin d’atteindre le meilleur résultat fonctionnel possible . Les résultats des PTG sont bons à excellents avec diminution des douleurs, amélioration des capacités fonctionnelles et de la qualité de vie . La longévité des PTG est bonne et dépasse 90 % de survie à dix ans .

Après chirurgie pour PTG, les soins de rééducation débutent dans le service de chirurgie orthopédique et se poursuivent dans la majorité des cas en hospitalisation pour rééducation ou moins fréquemment en hospitalisation de jour ou en externat directement à domicile. Récemment, Oldmeadow et al. ont développé et validé, en Australie, une méthode pour prédire les besoins d’hospitalisation en rééducation des patients après prothèse de hanche et de genou . Le score du Risk Assessment and Predictor Tool (RATP) identifie trois niveaux de risque avec une pertinence de 74,6 %. Quand le score RATP est inférieur à 6, la prédiction est d’hospitaliser et quand le score est supérieur à 9, la prédiction est un retour directement à domicile. Ce score n’est habituellement pas utilisé dans d’autres pays où les patients opérés d’une PTG sont adressés en centre de rééducation, même si cela ne correspond pas à une règle absolue. En fait, l’orientation des patients porteurs d’une PTG dépend de plusieurs facteurs comme l’âge, les comorbidités, les possibilités fonctionnelles postopératoires, les possibilités d’aide, les habitudes du chirurgien et les souhaits du patient . En France, le souhait du patient correspond à un critère d’aide à la décision mais n’est pas opposable au patient. Seules quelques études ont déterminé quels facteurs contribuaient à l’hospitalisation en rééducation en se focalisant sur la durée de séjour dans un but médicoéconomique sans prendre en considération les résultats fonctionnels à la sortie du centre de rééducation . L’objectif de cette étude a été de colliger les facteurs préopératoires qui ont déterminé la durée de séjour en hospitalisation pour rééducation en tenant compte des résultats des soins de rééducation définis par l’obtention d’une flexion active de genou de 90 degrés et d’un statut fonctionnel autorisant un retour directement à domicile.

2.2

Méthode

2.3

Résultats

Après sortie de l’étude de 38 patients qui ont présenté des complications, une cohorte de 244 patients opéré du genou a été caractérisée par les paramètres démographiques, les capacités fonctionnelles avant chirurgie, la présence ou non d’antécédent de comorbidités ou de prothèse articulaire ( Tableaux 1 et 2 ). Les patients ont été admis en hospitalisation pour rééducation en moyenne à 7,5 ± 1,7 jours de la chirurgie.

Tableau 1

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

Actualités : cinquième séminaire de la FATO Value and efficacy of early supported discharge from stroke units Treatment of muscle trauma in sportspeople (from injury on the field to resumption of the sport) Rehabilitation of arm function after stroke. Literature review Sacral reflexes and urinary incontinence in women: New concepts Evaluation of the effectiveness of botulinum toxin injections in the lower limb muscles of children with cerebral palsy. Preliminary prospective study of the advantages of ultrasound guidance

Stay updated, free articles. Join our Telegram channel

Join