1.1

Introduction

Knee osteoarthritis (OA) is the most common cause of knee joint cartilage wear. When medical treatment is not sufficient, surgery for total knee replacement (TKR) can improve pain, keep a good joint range of motion (ROM) and often patients recover functional gait abilities that improves their quality of life . According to the Knee Society, the results are considered positive when the knee is painless, stable in the frontal and anteroposterior directions (less than 5 mm), flexion reaches 110 degrees and flessum is less than 5 degrees. For functional abilities, the walking perimeter must be unlimited including walking up and down the stairs without technical aids . In France, these results were obtained thanks to the quality of surgery, anesthesia and pain management on top of rehabilitation care most often in a Physical Medicine and Rehabilitation (PM&R) center. For economic reasons and due to increasing demands for TKR surgery because of population aging, some alternatives to conventional hospital stays have been developed .

In France, some patients already have the opportunity to return home with rehabilitation care provided by a private physiotherapist. In other countries, like Australia, the Health System being different, the decision for patients to return home after TKR or total hip replacement (THR) is based on meeting presurgical criteria according to the Risk Assessment and Predictor Tool (RAPT) ( Risk Assessment and Predictor Tool described in Appendix 1 ) . A RAPT score less than 6 corresponds to a hospital stay after surgery; a RAPT score more than 9 allows the patient to return directly home whereas an intermediary RAPT score between 6 and 9 requires home care for this patient to return home. The patient’s wish must also be taken into account without this criterion being part of the predictive score. Thus, the RAPT has a good validity with correct predictability in 74.6% of the cases . This tool is very easy to implement yet it is not used in France for lack of validation in the French language.

Before looking at using the RAPT in France, the objective of the study was to validate if patients with complications after surgery were the ones identified at risk for complications. A retrospective search was conducted on the medical charts to identify, according to the RAPT score, the relative risk of complications before surgery for patients hospitalized in a rehabilitation unit or center after TKR surgery.

1.2

Method

1.3

Results

The analysis was based on 272 medical charts. The mean population age was 71.6 years ± 8 (46–87). Women were predominant (61.6%). The durations of stays in the surgical and PM&R units were respectively 8 days ± 3.7 (3–13 days) and 25.1 days ± 11.2 (6–123 days). Upon discharge from the PM&R unit the mean knee flexion was 97 degrees ± 12 and knee extension was 3 degrees ± 5. The mean walking perimeter was 356 meters ± 172 with the aid of two crutches in 90% of the cases.

The mean preoperative Lequesne functional pain score was 11.7 ± 3.2 (median 12, ranges 6–22). The mean RAPT score was 6.5 ± 2.3 (median 7, ranges 1–11).

According to the classification defined by the RAPT, only 33 patients (12.2%) could have returned home right away, one patient out of two (136) would have required homecare for a direct return home and 103 patients (37.8%) would have absolutely needed a hospital stay ( Table 1 ).

Gender repartition was significantly different according to the RAPT classification (chi ² : 78.4, P < 0.0001) with a ratio in favor of women if the RAPT score was less than 6, a ratio close to 1 if the RAPT score was between 6 and 9 and finally a ratio in favor of men if the RAPT score more than 9 ( Table 2 ). The age and preoperative Lequesne functional pain score were different according to the RAPT classifications. The patients that could have returned home directly (RAPT more than 9) were significantly younger by 3 to 8 years in average and had in average a 2-point lower preoperative Lequesne functional pain score. The duration of the PM&R stay was significantly longer with a mean of 4 days for patients who had a RAPT score less than 6 while knee mobility and walking perimeter upon discharge from the PM&R center were similar ( Table 3 ). The duration of the orthopedic surgical stay was also similar ( Table 3 ). Fifty-one complications (18.7%; CI 95%: 14–23.3%) challenged either the vital prognosis of the patient or the prosthesis’s viability. These complications were:

• •
  

  17 DVT (6.2%; CI 95%: 3.5–9.1%), including two that were associated to severe pulmonary embolism (PE);

  
  
• •
  

  14 urinary tract infections (5.1%; CI95%: 2.5–7.7%);

  
  
• •
  

  six knee stiffness treated by surgical mobilization under general anesthesia (2.2%; CI95%: 0.5–3.9%);

  
  
• •
  

  six wound skin surgeries (2.2%; CI95%: 0.5–3.9%);

  
  
• •
  

  six severe system failures due to preexisting pathologies that required patients to be transferred to a special unit (2.2%; CI95%: 0.5–3.9%) (twice for cardiac pathologies, twice for digestive pathologies, once for respiratory pathology and once for a non-infectious urinary pathology);

  
  
• •
  

  two deep infections of the TKR prosthesis (0.7%; CI95%: − 0.3–1.7%).

Patients that could have gone directly home had very few complications. The relative risk to have a complication was 0.45 meaning that a RAPT score more than 9 could be considered as a protective factor for the patient. Conversely, the risk of complication is high (relative risk at 2.16) when the patient has a RAPT score less than 6 ( Table 4 ).

1.4

Discussion

Orienting patients after a hospital stay in acute care is a healthcare challenge. TKR surgery is faced with this issue to optimize therapeutic care. The French Higher Health Authority (HAS) elaborated recommendations on January 2008 to answer this question . The objective was to: “propose guidelines to help the physician assess the need for the patient to go to a center for rehabilitation care after a hospital stay for acute care”. It was also recommended that, in light of organizational requirements and when possible, the decision should be made before the surgery by a multidisciplinary team taking into consideration several criteria: psychosocial, environmental, medical, clinical and functional elements. This approach is quite complicated to implement and has yet to be validated on a practical level. This is why Coudeyre et al. suggested implementing studies with good methodological quality in order to evaluate predictive orientation tools . We suggested using the RAPT as it was used in English-speaking countries but had very rarely been used in France . Oldmeadow et al. reported its simple use and validity in contrast with the complexity of the HAS preoperative recommendation criteria . However, this tool has not been translated into French and thus shows its validation limits in France. The US example reported that 78 to 96% of patients went directly home after 6 to 10 days in the hospital in the orthopedic surgical unit . In Australia, some authors showed that a shorter hospital stay did not lead to an increase in the number of complications or to subsequent hospital stays .

In light of these results, Oldmeadow et al. had the idea to develop a predictive index, the RAPT to orientate patients after surgery of the lower limb. In 2002, a first study, based on the results of three different hospitals, validated that 60% of patients went directly home after TKR surgery . These patients were young men with a good autonomy and easy access to rehabilitation care. In 2003, the second study validated the RAPT on a population of 520 patients with a mean age of 71.7 years . A model of logistic regression was build from nine identified factors.

Two factors were excluded since they were not significant: complications and home environment. The factor “patient’s wish” was also excluded since it could change according to the patient’s educational level even though it was very significant (odd ratio: 12.9; P : 0.0001). Six factors were finally kept in this model:

• •
  

  age (odd ratio: 0.95; P : 0.0001);

  
  
• •
  

  gender (odd ratio: 0.79; P : 0.18);

  
  
• •
  

  mobility (odd ratio: 1.67; P : 0.005);

  
  
• •
  

  home care (odd ratio: 1.67; P : 0.0001);

  
  
• •
  

  home help (odd ratio: 1.67; P : 0.27);

  
  
• •
  

  and above all access to PM&R care (odd ratio: 6.47; P : 0.0001).

The model was exact in 75.2% of the cases. Based on the significance of the various factors, the RAPT score was categorized into three classes (less than 6; 6–9 and more than 9). RAPT validation was then determined by a prospective analysis on a population of 130 subjects similar to the previous population but younger (68.7 years ± 9.5). The orientation of patients was accurate in respectively 89.2, 62.3 and 83.7% of the cases according to the three classes of the RAPT score. The greater limit of this predictive model was reached with the unknown proportion of patients who had TKR or THR. A third article, published in 2004, focused on two cohorts of 50 patients who had TKR or THR surgery (proportion of two-thirds) . The first cohort with a mean age of 69.9 had surgery during the year 2000 and was the object of a retrospective study. The second cohort, mean age 66.7 years, had surgery in 2001 and was oriented according to the RAPT score.

The results showed that in 2000 only 34% of patients went directly home vs. 64% of patients in 2001. The number of subsequent hospital stays within the year was not different in 2000 and in 2001 (4 and 6%) and almost exclusively concerned patients who had a RAPT score between 6 and 9. The reasons for these subsequent hospital stays for patients who had TKR surgery were linked to complications: two surgical knee mobilizations under general anesthesia (10%) in 2000 vs. only one in 2001 (4.7%) and one TKR infection in 2001(4.7%).

The results from this study showed that patients with a RAPT score less than 6 had a high risk of complications whereas patients with a RAPT score more than 9 had a low risk of complications. Before generalizing our study’s results according to the RAPT score, it was necessary to know if the number of complications we reported was similar to the one described in the literature.

Mobilization under general anesthesia for knee stiffness was necessary in 2.2% of cases in our population a percentage comparable to the data found in the literature ranging from 2 to 4.75% . This treatment affected one patient out of 33 when the RAPT score less than 6; one patient out of 136 when the RAPT score more than 9 and four out of 103 when the RAPT score was between 6 and 9.

A greater number of patients, around 10%, could have been concerned if threshold was set at 80 or 90 degrees of TKR flexion as suggested by some authors .

We reported a 6.2% rate of DVT, which seems high compared to the 1% rate showed by Seah et al. . This could be explained by our older population compared to the Anglo-Saxon populations (mean age of 65 years). However, Hitos et al. reported a higher incidence of 5.7% (95% CI: 1.6–18.6%) on a 3-month postoperative follow-up . During the systematic search for DVT with a Doppler ultrasound, Schellong et al. reported a diagnosis rate of 11.5% for asymptomatic DVT in the 7 days post-TKR or THR surgery . This rate was even higher after contrast peripheral Phlebography (18.9%). In this study, only the most vulnerable patients with a RAPT score less than 6 were at risk (RR: 3).

The incidence rate of TKR deep infections was comparable to the data found in the literature with 0.7% vs. 0.2 to 3.9% . No specific risk was unveiled by the RAPT. Knee wound healing delay and UTIs are problematic since there is a contamination risk for the prosthesis. There is a risk factor for patients with a RAPT score less than 6 and a specific monitoring with UTI screening should be implemented especially for patients with a RAPT score between 6 and 9 (RR: 1.31). This study’s result of 5.1% of UTIs is similar to the one reported by Lingaraj et al. with 8% patients with UTIs (mean age 67.5) after TKR surgery .

Regarding system failure due to preexisting conditions, they were more common in the most vulnerable patients with a relative risk at 3.45.

Furthermore, all these complications were observed for similar durations of stay in the orthopedic surgical unit. The duration of stay in the PM&R unit was 4 days longer for the most vulnerable patients who had a preoperative RAPT score less than 6. This result was explained by the complications observed, given similar knee amplitudes and gait functions, upon leaving the PM&R unit.

1.5

Conclusion

Few patients hospitalized in a PM&R center for post-TKR rehabilitation had the possibility to return directly home with a RAPT score more than 9 (12.2%). These patients with few complications were characterized by their young age and the fact that they were all men. Nevertheless, according to our study they spent an average of 23 days in a PM&R center to regain proper knee mobility. On the contrary, patients who had the most complications and required a longer hospital stay were identified as the most vulnerable according to the RAPT (RAPT less than 6). Patients with a RAPT score between 6 and 9 did not have a high risk of complications. Setting up social help support before surgery could be interesting so these patients wouldn’t have to stay in a PM&R center. From these results it would be relevant to conduct a French prospective multicenter study to test the use of the RAPT in order to orientate patients before TKR surgery in order to know more about their relative risks.

Disclosure of interest

The authors declare that they have no conflicts of interest concerning this article.

2.1

Introduction

L’arthrose représente la cause la plus fréquente de destruction de l’articulation du genou. Lorsque le traitement médical est dépassé, la mise en place chirurgicale d’une prothèse de genou (PTG) permet d’améliorer les douleurs, de conserver une bonne mobilité articulaire et le plus souvent de retrouver une activité fonctionnelle de déplacement qui préserve la qualité de vie . Selon la Knee Society, les résultats sont considérés comme très bons quand le genou est indolore, stable dans le plan frontal et antéro-postérieur (moins de 5 mm), que la flexion atteint 110 degrés et que le flessum est inférieur à 5 degrés. Sur le plan des capacités fonctionnelles, le périmètre de marche doit être illimité avec montée et descente des escaliers sans aide technique . En France, ces résultats sont obtenus grâce à la qualité de la chirurgie, de l’anesthésie et du contrôle de la douleur, en plus des soins de rééducation qui sont dispensés le plus souvent en hospitalisation dans un centre de médecine physique et réadaptation (MPR). Pour des raisons économiques et du fait d’une demande grandissante de mise en place de PTG en raison du vieillissement de la population, les alternatives à l’hospitalisation conventionnelle se sont développées . Certains chirurgiens proposent déjà à leurs patients un retour directement à domicile avec soins de masso-kinésithérapie en secteur libéral. Dans d’autres pays, comme l’Australie, le système de santé étant différent, le retour directement à domicile des patients après PTG ou prothèse totale de hanche (PTH) est prônée après étude de critères pré-chirurgicaux établis selon l’indice Risk Assessment and Predictor Tool (RAPT) ( Risk Assessment and Predictor Tool en Annexe 1 ) . L’indice RAPT inférieur à 6 correspond à une hospitalisation, l’indice RAPT supérieur à 9 permet un retour directement à domicile alors que l’indice RAPT intermédiaire compris entre 6 et 9 nécessite la mise en place d’aides à domicile pour un retour à domicile. Le souhait du patient doit également être pris en considération mais sans que ce critère ne soit pris en compte par le score de prédiction. Ainsi, l’indice RAPT présente une bonne validité avec une prévision correcte dans 74,6 % des cas . Il est très simple à mettre en pratique alors qu’il n’est pas utilisé en France car non validé en langue française. Aussi avant d’envisager son utilisation dans notre pays, l’objectif de ce travail a été de savoir si les patients qui présentaient des complications étaient ceux qui étaient identifiés comme à risque par l’indice RAPT. Il a donc été recherché à postériori le risque relatif de complication en fonction du score RAPT présenté en préopératoire par les patients hospitalisés pour rééducation après pose d’une prothèse totale de genou.

2.2

Méthode

2.3

Résultats

Deux cent soixante-douze dossiers ont été analysés. L’âge moyen de la population était de 71,6 ans ± 8 (46–87) avec une prédominance de femme (61,6 %). Les durées de séjour en service de chirurgie et de MPR étaient respectivement de huit jours ± 3,7 (3–13 jours) et de 25,1 jours ± 11,2 (6–123 jours). La flexion et l’extension moyenne des genoux à la sortie du service de MPR a été de 97 degrés ± 12 et de 3 degrés ± 5 respectivement. Le périmètre de marche a été de 356 mètres ± 172 avec port de deux cannes anglaises dans 90 % des cas. L’indice algo-fonctionnel moyen de Lequesne préopératoire était de 11,7 ± 3,2 (médiane 12, extrêmes de 6 à 22). L’indice RAPT était de 6,5 ± 2,3 (médiane 7, extrêmes de 1 à 11).

Selon la classification définie par l’indice RAPT, seulement 33 patients (12,2 %) auraient pu rentrer directement à leur domicile, un patient sur deux (136) auraient nécessité de mettre en place des aides pour un retour à domicile directement et 103 patients (37,8 %) auraient nécessité absolument d’une hospitalisation ( Tableau 1 ).

Tableau 1

Répartition des patients en fonction de l’indice RAPT.

Indice RAPT	n	%	% Cumulé	Classe RAPT
12	0	0	0	12,2 %
11	7	2,6	2,6
10	26	9,6	12,2
9	30	11	23,2	50 %
8	38	14	37,2
7	36	13,2	50,4
6	32	11,8	62,2
5	41	15,1	77,3	37,8 %
4	33	12,1	89,4
3	17	6,3	95,7
2	11	4	99,7
1	1	0,4	100

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:

Compte rendu du 26 econgrès de la SOFMER à Nantes Isolated greater tuberosity fracture: Short-term functional outcome following a specific rehabilitation program Spontaneous confabulations and behavioral and cognitive dysexecutive syndrome Unusual clinical presentation of a partial tibialis anterior rupture Acquired visual field defects rehabilitation: Critical review and perspectives Evaluation of medical students’ expectations for multimedia teaching materials: Illustration by an original method using the evaluation of a web site on cardiovascular rehabilitation

Stay updated, free articles. Join our Telegram channel

Join