A prospective randomized case-control study was done from November 2017 to July 2018. An institutional review board approval was obtained for our research protocol to prospective data acquisition of patients undergoing TKA (CAAE:11677714.4.0000.5149), and an informed consent was signed by all the participants or by a person responsible for them.
The sample size was calculated to test the hypothesis that a differentiated educational program would increase post-operative range-of-motion (primary outcome). Trying to detect a difference of 10 degrees in range-of-motion (ROM) between groups and based on a power test of 80% and a confidence interval of 5% after assuming a possible 10% loss in follow-up, 15 patients in each group was considered the minimum number of participants. SF-36 and WOMAC results were considered secondary outcomes.
Inclusion criteria: patients with unilateral symptomatic primary or secondary OA of the knee, older than 45 years, with indication of a primary TKA, who signed the informed consent form. Immediately after sign the inform consent, pre-operative physical examination and functional tests were applied to all participants by two investigators (DGKB, DSL). Patients were then randomized into two groups (intervention or control), by choosing one of two closed envelopes in which was written the words conventional or differentiated, referring to the approach it would be done. The post-operative clinical and functional evaluation was performed by three blinded investigators (MVTR, TFGM, FSM). The study was conducted at the Knee Surgery facility of Hospital das Clinicas da Universidade Federal de Minas Gerais.
In the control group, a verbal education about TKA was made during the preoperative appointments as done usually at the Institution. In the intervention group, after the usual verbal education, patients received a differentiated orientation that included discussion and lectures done by an orthopedic surgeon, a nurse and a physiotherapist concerning the pre and post-operative care, pain management, rehabilitation exercises to gain ROM, the basic steps of the surgical procedure and the importance of walking with a walker. FIGURE 1
Seventy-nine patients were eligible for the study. Twelve were excluded during follow-up: one patient died with a heart attack, three patients had periprosthetic joint infection and eight did not return for control and were excluded (FIGURE 2).
The two groups were comparable regarding demographics, clinical and functional tests. (Table 1)
Table 1
– Demographics between groups before study enrollment
| Control | Intervention |
n | 29 | 38 |
Female | 23 (79.3%) | 30 (78.9%) |
Age* | 62.8 (11.8) | 65,1 (9.3) |
BMI* | 29,2 (5.5) | 30,2 (5.5) |
Secondary OA | 6 (20.7%) | 7 (18.4%) |
Right Knee | 11 (37.9%) | 17 (44.7%) |
Varus Deformity | 22 (75.9%) | 29 (76.3%) |
Presence of Comorbiditiesβ | 7 (24.1%) | 6 (15.8%) |
Analog Pain Scale* | 7.6 (2.0) | 7.6 (2.0) |
ROM* | 104.6 (25.5) | 103.9 (22.7) |
WOMAC* | 34.6 (16.0) | 37.3 (21.6) |
SF36* | 84.9 (13.8) | 85.3 (14.6) |
Walk > 400 m | 10 (34.4%) | 13 (34.2%) |
With Stair Impairment | 28 (96.6%) | 37 (94.7%) |
Use of walkers, canes | 10(34.5%) | 16(42.1%) |
* Mean and standard deviation |
β comorbidities means diabetes mellitus and/or arterial blood hypertension |
All patients were operated on in the same Institution by three different surgeons (MAPA, GMAS, TVOC) following the same protocol. Patients received a peridural anesthesia with bupivacaine (0,5%) and intravenous sedation (Diprovan®- AstraZeneca). Tourniquet was applied in all cases set to 300 mmHg. Primary TKA was performed through a classical medial arthrotomy with patellar eversion. A cruciate-retained implant (Nexgen® Zimmer – Warsaw, IN), with patellar substitution, fixed with a non-impregnated antibiotic cement, was used in all cases.
Postoperative multimodal pain control protocol was made in both groups and started on the same day of the procedure. It consisted of scheduled acetaminophen 750 mg every 6 hours, metamizole 500 mg every 6 hours, tramadol 50 mg every 8 hours for 24 hours and morphine 2 to 6 mg every 4 hours, as needed. Patients were encouraged to start early ROM and weight bearing with a walker in the first day. Patients were discharged from hospital on day 2 or 3.
Statistical analysis was performed to determine statistically significant differences between the 2 groups (p < 0.05), using appropriate software (G*Power Version 3.1.9.2) and included analysis of distribution by Zolmogorov-Smirnov test. Independent T-test was used to analyze numerical, continuous and normally distributed variable. Qui-square test was used for categorical data.