Evaluation of input data modality choices on functional gene embeddings.

Functional gene embeddings, numerical vectors capturing gene function, provide a promising way to integrate functional gene information into machine learning models. These embeddings are learnt by applying self-supervised machine-learning algorithms on various data types including quantitative omics measurements, protein-protein interaction networks and literature. However, downstream evaluations comparing alternative data modalities used to construct functional gene embeddings have been lacking. Here we benchmarked functional gene embeddings obtained from various data modalities for predicting disease-gene lists, cancer drivers, phenotype-gene associations and scores from genome-wide association studies. Off-the-shelf predictors trained on precomputed embeddings matched or outperformed dedicated state-of-the-art predictors, demonstrating their high utility. Embeddings based on literature and protein-protein interactions inferred from low-throughput experiments outperformed embeddings derived from genome-wide experimental data (transcriptomics, deletion screens and protein sequence) when predicting curated gene lists. In contrast, they did not perform better when predicting genome-wide association signals and were biased towards highly-studied genes. These results indicate that embeddings derived from literature and low-throughput experiments appear favourable in many existing benchmarks because they are biased towards well-studied genes and should therefore be considered with caution. Altogether, our study and precomputed embeddings will facilitate the development of machine-learning models in genetics and related fields.

A systematic benchmark of machine learning methods for protein-RNA interaction prediction.

RNA-binding proteins (RBPs) are central actors of RNA post-transcriptional regulation. Experiments to profile-binding sites of RBPs in vivo are limited to transcripts expressed in the experimental cell type, creating the need for computational methods to infer missing binding information. While numerous machine-learning based methods have been developed for this task, their use of heterogeneous training and evaluation datasets across different sets of RBPs and CLIP-seq protocols makes a direct comparison of their performance difficult. Here, we compile a set of 37 machine learning (primarily deep learning) methods for in vivo RBP-RNA interaction prediction and systematically benchmark a subset of 11 representative methods across hundreds of CLIP-seq datasets and RBPs. Using homogenized sample pre-processing and two negative-class sample generation strategies, we evaluate methods in terms of predictive performance and assess the impact of neural network architectures and input modalities on model performance. We believe that this study will not only enable researchers to choose the optimal prediction method for their tasks at hand, but also aid method developers in developing novel, high-performing methods by introducing a standardized framework for their evaluation.

Discrepant histological diagnoses: A cause of early low FJS-12 score and if untreated, unhappy Total Knee Arthroplasty patient.

PURPOSE: Total Knee Arthroplasty (TKA) is one of the most successful operations in orthopedics. Still, a sizable percentage of patients (20%) remain dissatisfied after a well-executed TKA. The study aims to examine the excised synovium from the suprapatellar region in osteoarthritic knees during TKA and evaluate the histopathology (HP) report to know whether discrepant diagnoses affect the Forgotten Joint Score-12 at various time intervals. METHODS: This is a prospective cohort study. Two hundred (160 female; 40 male) end-stage osteoarthritis patients who underwent primary TKA were studied. An inclusion criterion was patient with end-stage osteoarthritis. Clinically and serologically proven rheumatoid arthritis patients were excluded from the study. The synovium excised during the TKA procedure was sent for the HP examination. The statistical significance was measured with the Chi-square test and two-sample t-test. RESULTS: A total of 184 out of the 200 patients (92%) knee synovium showed HP features of osteoarthritis. The discordant diagnoses and discrepant diagnosis rate was 8% and 7%, respectively, which is statistically significant by Chi-square test (p value < 0.0001 and p value = 0.0001). 14 of the patients (12 F:2 M) showed histological features of inflammatory/rheumatoid arthritis who were treated, two patients (all female) showed HP features of villonodular synovitis. The mean (SD) improvement in FJS-12 at six weeks in the concordant group (25.3 [17.6]) is significantly more than the discrepant group (15.3 [12.5]), p-value 0.0385. CONCLUSION: 8% of our patients exhibited unexpected results. The study showed a 7% rate of discrepant diagnosis. This discrepant diagnosis if missed and untreated, would have affected the function and long-term survival of the implanted TKA.

The influence of the internet on the patients' choice of surgeon for their total knee replacement surgery.

BACKGROUND: To study the influence of the internet on the patient's choice about Total Knee Replacement surgery and about their choice of surgeon for their Total Knee Replacement procedure. SIGNIFICANCE: Insight into patient's decision making will help the surgeon to increase his/her visibility and also will help in providing true and reliable information about the patient's ailment. MATERIALS AND METHODS: We studied 200 patients who were operated by a single surgeon. They were given a questionnaire which had various questions regarding their name, age, sex, date of operation. It also had leading questions like did you or your relative searched the net about the operation and the operating surgeon? The information was collected by self filled questionnaire form. RESULTS: There were 166 females and 34 males. 70% of to be operated patients or their relatives searched the net before undergoing the Total knee replacement. Also 36% of patient relatives (mainly son or daughter) searched the surgeon on the net before finalizing the operation. 10% of patients searched the operating surgeon on the internet. 15% of patients or their relatives chose the operating surgeon based on their net search. CONCLUSION: Our study shows that a sizable number of decision makers (patients and their children) search the information about the operation and about their operating surgeon. Hence it is recommended that the surgeon should have an information portal to increase his/her visibility, patient reach and imparting proper information to the prospective patients regarding the operative procedure.

Education, engagement and provision of empathy by trained counselor enhances the patient satisfaction after Total Knee Arthroplasty.

BACKGROUND: One of the major concerns following Total Knee Arthroplasty (TKA) surgery is patient satisfaction. Hence, this study sought to assess the effect of patient education, engagement and provision of empathy delivered by trained counselor on patient satisfaction. METHODS: All the patients included in the study were randomly allocated to two groups: Patients who received counseling from trained personnel during pre-, peri- and post-TKA phase constituted group-A (n = 100) whereas patients who were advised by the surgeon before the procedure constituted group-B (n = 100). All the patients who were counseled by trained personnel continued to receive counseling up to six months following TKA surgery. Patients belonging to group-A were provided counseling pertaining to education related to TKA procedure, the modalities of pain management and post-TKA ambulation regimen. Patient satisfaction was measured with Forgotten Joint Score-12 (FJS-12) and consumer gap (defined as the difference between patient perception and expectation) was recorded by an independent observer after 52-week of the TKA procedure. RESULTS: Patients belong to group-A had higher satisfaction as compared to those in group-B as evident by statistically higher FJS-12 score (64.5 ± 16.4 vs. 59.4 ± 17.9; p = 0.0369). There was statistically significant difference observed in consumer gap between group-A (6%) and group-B (16%) (p = 0.0242). CONCLUSIONS: The results of the study are encouraging to employ counseling by trained personnel during pre-, peri- and post-TKA phases as to improve patient satisfaction and reduce consumer gap.

Comparison of local pain and tissue reaction between conventional pneumatic tourniquet and disposable silicone ring tourniquet during Total Knee Arthroplasty.

BACKGROUND: To study the comparison of conventional pneumatic and disposable silicone ring tourniquet in Total Knee Arthroplasty. MATERIAL AND METHODS: This is a prospective randomized control trial. We used conventional pneumatic tourniquet on one side of leg and disposable silicone ring tourniquet on the other side in consecutive 50 simultaneous bilateral TKR patients. Patients having peripheral vascular disease of the lower limb were excluded from the study. The patient demographics & characteristics are identical being the same patient with two different legs. We started the study with null hypothesis. An independent observer assessed the local tourniquet site pain (VAS score 1-10) and local tourniquet site skin reaction at 24 h and 48 h after the TKA. P value < 0.05 was considered significant. RESULTS: There were no local skin complication with disposable tourniquet (0%). 8 out of 50 patients in whom the conventional tourniquet was applied showed local bruising, and two patients had blister formation making the local skin site complication rate 20% (statistically significant, p value0.0196, chi-squared test). The VAS score at 24 h was 4.3 ± 1.5 for disposable tourniquet group as against 5.6 ± 2.1 for conventional tourniquet group (statistically significant, p value = 0.0152, t statistic test for comparison of means). The VAS score at 48 h was 2.1 ± 1.5 and 3.3 ± 1.2 for disposable tourniquet group and conventional tourniquet group respectively (statistically significant p value = 0.003, student's t-test). CONCLUSION: Use of disposable tourniquet has better outcome than the conventional tourniquet with minimal or no local complications. The advantages of the disposable tourniquet are: 1 less local pain, 2. no local skin problems, 3. accurate tourniquet pressure at the application site, 4.0% local contamination. Hence, we recommend use of the disposable tourniquet during the Total Knee Arthroplasty.

Study of Patients with Bilateral Knee Osteoarthritis Undergoing Total Knee Replacement Procedure with Coexisting Lumbar Spondylosis Symptoms.

STUDY DESIGN: This is a prospective cohort study involving patients who were followed for 2 years after total knee replacement (TKR) to determine changes in lumbar spine and knee symptoms. PURPOSE: The objectives of this study were to determine the percentage of patients undergoing bilateral TKR who present with coexisting lumbar spine problems and determine if TKR relieves lumbar spine symptoms. OVERVIEW OF LITERATURE: No studies quantify the percentage of TKR patients who experience relief of lumbar spine symptoms after TKR surgery. METHODS: The study included 200 patients (164 females, 36 males) undergoing primary TKR. Follow-up was performed at 4 weeks, 3, 6, 12, and 24 months. Lumbar spine and knee symptom improvements were assessed using the Oswestry Disability Index (ODI) and Oxford Knee Score, respectively. RESULTS: All 200 patients undergoing bilateral TKR presented with radiographic lumbar spine degenerative pathology; 60% (n=120) of the patients presented with moderate to severe clinical symptoms of lumbar spondylosis, including 54% (n=108) with degenerative lumbar spondylosis and lumbar canal stenosis and 6% (n=12) with degenerative spondylolisthesis. Of the 120 patients who presented with lumbar spine problems, 90% (n=108) reported improvement in their symptoms; the ODI score improved from 42.5%±4.1% preoperative score to 15.6%±2.3% postoperative score (p-value<0.001). Of the 12 patients with no improvement, 10 patients underwent percutaneous procedures for their lumbar spine pathology with good results, one patient underwent surgery, and one declined any intervention. CONCLUSIONS: A significant number of patients (60%) undergoing bilateral TKR also present with symptomatic lumbar spine problems. Patients with mild to moderate lumbar spine degenerative symptoms and no associated severe radiating pain on activity are more likely to experience relief of their symptoms post-TKR.

Home physiotherapy with vs. without supervision of physiotherapist for assessing manipulation under anaesthesia after total knee arthroplasty.

The aim of this retrospective cohort study was to compare home physiotherapy with or without supervision of physiotherapist for assessing manipulation under anaesthesia after total knee arthroplasty. METHODS: A total of 900 patients (including 810 females and 90 males) who had undergone total knee arthroplasty were divided into group A (n = 300) and group B (n = 600). Patients in group A had home physiotherapy on their own after discharge from hospital. The physiotherapist did not visit them at home. Patients in group B received home physiotherapy under supervision of physiotherapist for 6 weeks after discharge from hospital. Patients' age, range of motion of the knee, and forgotten joint score-12 were assessed. A p < 0.05 was considered statistically significant. RESULTS: In group A, the mean age was 69.1 ± 14.3 years (range: 58 to 82 years); in group B, the mean age was 66.5 ± 15.7 years (range: 56 to 83 years) (p > 0.05). Preoperatively, the mean range of motion of the knee in group A and B was 95.8° ± 18.1° and 95.4° ± 17.8°, respectively (p > 0.05). The mean forgotten joint score-12 of group A and B were 11.90 ± 11.3 and 11.72 ± 12.1 (p > 0.05), respectively. Six weeks after total knee arthroplasty, the mean ROM of the knee in group A and B was 109.7° ± 22.3° and 121° ± 21.5°, respectively (p < 0.05). The mean postoperative forgotten joint score-12 of the group A and B was 24.5 ± 16.4 and 25.6 ± 17.4, respectively (p > 0.05). The rate of manipulation under anaesthesia was 3% in group A and 0.2% in group B (p < 0.05). CONCLUSION: After total knee arthroplasty, frequent physiotherapist's instruction helps the patients improve knee exercises and therefore decrease the risk of revision surgery. The home physiotherapy under supervision of physiotherapist lowers the rate of manipulation under anaesthesia. LEVEL OF EVIDENCE: Therapeutic study, Level IIa.

Understanding the apprehension and concern haunting patients before a total knee arthroplasty.

PURPOSE: The aim of this prospective study was to understand the fear and apprehension factors that play on patient's mind before total knee arthroplasty. METHODS: This retrospective cohort study included 500 consecutive patients (375 females and 125 males) who were scheduled to undergo total knee arthroplasty the next day. The patients were asked to list the most important fear in their mind regarding the operation in descending order of importance. They were given a questionnaire form which contained several capture points, including age, gender, educational background, occupation, and provision of help at home. Preoperative pain was measured by using the visual analog score, and its influence on the patients' fear and apprehension factors was also measured. RESULTS: In this study, 58% of patients (50 males, 40%; 240 females, 64%) were fearful of the pain that they would experience after surgery and during the postoperative physiotherapy. The female patients showed more fear of pain than their male counterparts (P < 0.05). 18% of the patients (40 males, 32%; 50 females, 13%) listed whether they will be able to walk and perform activities of daily living after surgery as the most important fear. The male patients had more fear of returning to normal walking (P < 0.05). 20% of the patients (30 males, 24%; 70 females, 19%) were fearful about getting adequate home help after discharge from hospital (P > 0.05). 4% of patients were concerned about withstanding such a major operation. There was no difference between male and female patients (P > 0.05). CONCLUSION: The majority of the patients experience apprehension of pain in the perioperative period of TKA. Preoperative counseling benefits pain management by alleviating the patient's concerns about the fear of postoperative pain and apprehension of returning to normal walking.

CRP (C Reactive Protein) level after total knee replacement in Indian population--- does it follow Anglo-Saxon trend?

BACKGROUND: This study was to determine how C-reactive protein (CRP) responds after total knee replacement (TKR), including both unilateral and simultaneous bilateral TKR in Indian population and if it follows Anglo-Saxon trend. Published literature from North America and Europe shows CRP value peaks on the 2nd post-operative day and drops to normal by 6-8 weeks. We started the study with null hypothesis. MATERIAL AND METHODS: This is a prospective study, with 50 patients (all females, 25 received unilateral operations and 25 bilateral ones) included. CRP levels were measured, on the 2nd day, 8, 12 and 16 weeks after operation. RESULTS: In both groups, CRP level rose the 2nd post-operative day. The rise in CRP level was significantly higher in the simultaneous bilateral TKR group than in the unilateral TKR group. In unilateral cases, CRP on the 2nd postoperative day ranged from 65 to 110 mg/l with average level of 80 mg/ml. In bilateral TKR cases, CRP level on the 2nd postoperative day was between 110 and 180 mg/l with a mean of 140 mg/ml. The CRP level returned to normal in about 40% of unilateral TKR patients 8 weeks after operation, while in 92% (23 out of 25) of bilateral simultaneous TKR patients it stayed at a high level 8 weeks post-op and did not come back to normal. At 12 weeks CRP decreased to normal in all 100% of unilateral TKR patients and 32% of bilateral TKR patients. At 16 weeks, CRP was normal in all bilateral TKR patients. CONCLUSION: 60% of our unilateral TKR patients and 92% of our simultaneous bilateral TKR patients did not achieve a normal CRP 8 weeks after operation. These findings are significant as CRP is commonly used as a very sensitive indicator of postoperative joint infection. Hence we conclude that in the Indian TKR patients the CRP values take longer time to return to normal than in their Anglo-Saxon counterparts. Published results regarding the normal levels of CRP in unilateral TKR should not be extrapolated to simultaneous bilateral TKR patients.