It was a prospective study conducted between May 2016 to April 2017 in a tertiary health facility in Southwest Nigeria. Ethical approval was obtained from the Institution’s Ethical Committee (IRB/IEC/004553 NATIONAL:NHREC/27/02/2009a) while written consent was obtained from the patients. The study was done in compliance with ethical standard laid down in 1964 Declaration of Helsinki and its later amendments or comparable ethical standards. Forty-five consecutive patients who met the inclusion criteria were recruited into the study. These are the patient with mild/moderate knee osteoarthritis who were willing to receive either of the treatments. Patients with rheumatoid arthritis, post traumatic osteoarthritis, pregnancy, thrombocytopenia were excluded from the study.

Detailed patient assessment was carried out including the clinical history, examination, full blood count, plain radiograph of the knee. Patients were categorised into mild/moderate osteoarthritis using the Kellgreen Lawrence grading [15] and outcome measures of the treatment were evaluated using WOMAC and VAS [16,17]. Scoring was done by the same individual.

Consecutive patients were serially recruited into 3 Groups (A,B,C). Group A had only intra articular PRP while Group B had oral NSAID and Group C had a combination of oral NSAID (Diclofenac preparation) and intra articular PRP.

PRP was prepared from 20 mL of whole blood obtained from the ante-cubital vein using a sterile bottle preloaded with citrate phosphate dextrose-adenine. This was initially centrifuged at 1,800 RPM for 15 minutes following which the supernatant was obtained and further subjected to centrifugation at 2,500 RPM for a further 15 minutes to separate the plasma and platelet. The PRP obtained were randomly subjected to platelet count using auto analyser to be sure the correct concentration of platelet was present. Three millilitres of suspended platelet rich plasma were administered into the joint using standard landmark. This procedure was repeated at 4 weekly intervals for maximum of 3 doses. Patients in Group C had in addition to the injection, oral administration of diclofenac up to the end of the intervention phase of the study. Patients randomised into Group B had only oral diclofenac (arthrotec75 mg 12 hourly) which was administered as necessary, but not more than twice in a day.

Potential confounders such as therapy that can influence the study (pain) were disallowed as patients that required further medication or therapy were excluded from the study. Subjects were followed-up on monthly basis for 6 months with routine evaluation of the patient using WOMAC questionnaire and VAS score.

Data were analysed using the IBM SPSS Statistics Version 22. Armonk, NY; IBM Corp. Data were expressed as mean unless otherwise indicated. Mean and standard deviation of the WOMAC, VAS was calculated. Comparison of the means was done using analysis of variance (ANOVA).

Forty-five patients were recruited into three groups. The mean age of the population was 60.2±11.7 years with a range of 41 years to 85 years. Sixty percent of the subjects had bilateral involvement of the knee while the rest had unilateral involvement [Table/Fig-1].

All the three groups experienced an improvement in a similar trend with the commencement of the treatment. At the commencement of the study, all the groups had similar mean VAS score (p=0.630) but at the end of this study, the mean VAS were 1.46 and 1.66, respectively in Group A and C while Group B had 4.26 and this was statistically significant (p=0.001)

Significant differences in the VAS score were found in the 5^th month, as the groups that had only NSAID was found to have a statistically significant VAS than groups who had a form of intra-articular administration of PRP [Table/Fig-2].

There was an initial improvement in symptoms up to the 4^th month after which there was a significant difference in the mean WOMAC score of the group of patient who had only NSAID (Group B) and those who had any form of PRP administration.

At the 6^th month, the mean WOMAC score were 13.40 and 22.23, respectively in the Group A and C while Group B was 44.0 which was found to be significantly greater than the other groups (p=0.001) [Table/Fig-3].

Early outcome measures (VAS and WOMAC) of this study revealed that intra-articular platelet rich plasma administration in the knee joint has the capacity to improve the symptoms experienced by patient with mild and moderate knee osteoarthritis immediately and up to 3 months after the last administration.

However, the result obtained in Group A (PRP only) cannot only be explained by the chondrogenesis theory which is expected to take effect much later. It is likely that the effects of PRP in the knee joint is beyond chondrogenesis as revealed by early resolution of pain and improved activity level in this study. Investigators have proposed that PRP is likely to have a whole lot of wider impact on the joint beyond chondrogenesis proposing that other effects could be due to reduction of synovial membrane hyperplasia, cytokine modulation which may be responsible for the early effects while other factors will be responsible for the pain controlling effect subsequently [9,18,19].

The present authors found that all patients experienced an improvement in their symptoms within the 1^st month of receiving PRP injection in the present series. Patel S et al., also found the mean duration of benefit after the 1^st injection was 17.63 days [18]. This further lay credence to the fact that chondral remodelling which would have started later and lasted longer could not solely explain the effects of PRP.

The improvement in the VAS and WOMAC observed within the first three months of platelet rich plasma intra-articular could suggest that it may have an immediate action on the joint which is comparable in efficacy with the NSAIDs but with the advantage of an extended action lasting up to the 6^th month of the study.

The present study found that patients in PRP group had significant alleviation of their symptoms compared to those in NSAID. A similar outcome was found in a similar study that compared PRP with Acetaminophen where subjects revealed a significantly better outcome in the PRP group for up to 6 months [20]. The combination of platelet rich plasma and NSAIDs as a form of therapy in the management of osteoarthritis have not been extensively studied and it has been postulated that the anti-inflammatory actions of NSAIDs may impair the inflammatory stage of cartilage repair that is meant to be an effect of the platelet rich plasma. This may be the reason why we found a better response both to pain and function in PRP group only compared to Group C subjects (PRP combined with NSAID) though this was not statistically significant. It is also believed that PRP has inhibitory effect on joint inflammation by acting against NF-kβ pathway [21]. Hence combination of both is expected to be synergistic which we did not find in the present study.

The authors acknowledge the fact that the present study was limited to mild/moderate cases of osteoarthritis as authors only recruited Grades 1 and 2 using the Kellgren-Lawrence system. The follow-up period was short therefore a study that will consider the above limitation will help in further shedding light to this treatment modality.

In conclusion, the present authors found early efficacy following intra-articular platelet rich plasma administration in the treatment of mild and moderate knee plasma regardless of its combination with NSAID. It should be offered as one of the first line of treatment in indicated patients giving its immediate effect on the patients’ symptoms.