Prediction of postpartum hemorrhage (PPH) using machine learning algorithms in a Kenyan population.

Introduction: Postpartum hemorrhage (PPH) is a significant cause of maternal mortality worldwide, particularly in low- and middle-income countries. It is essential to develop effective prediction models to identify women at risk of PPH and implement appropriate interventions to reduce maternal morbidity and mortality. This study aims to predict the occurrence of postpartum hemorrhage using machine learning models based on antenatal, intrapartum, and postnatal visit data obtained from the Kenya Antenatal and Postnatal Care Research Collective cohort. Method: Four machine learning models - logistic regression, naïve Bayes, decision tree, and random forest - were constructed using 67% training data (1,056/1,576). The training data was further split into 67% for model building and 33% cross validation. Once the models are built, the remaining 33% (520/1,576) independent test data was used for external validation to confirm the models' performance. Models were fine-tuned using feature selection through extra tree classifier technique. Model performance was assessed using accuracy, sensitivity, and area under the curve (AUC) of the receiver operating characteristics (ROC) curve. Result: The naïve Bayes model performed best with 0.95 accuracy, 0.97 specificity, and 0.76 AUC. Seven factors (anemia, limited prenatal care, hemoglobin concentrations, signs of pallor at intrapartum, intrapartum systolic blood pressure, intrapartum diastolic blood pressure, and intrapartum respiratory rate) were associated with PPH prediction in Kenyan population. Discussion: This study demonstrates the potential of machine learning models in predicting PPH in the Kenyan population. Future studies with larger datasets and more PPH cases should be conducted to improve prediction performance of machine learning model. Such prediction algorithms would immensely help to construct a personalized obstetric path for each pregnant patient, improve resource allocation, and reduce maternal mortality and morbidity.

Long-term Wear Data From a Prospective Multicenter Study of Second-Generation Highly Cross-linked Polyethylene Inserts in Total Hip Arthroplasty.

In a prospective multicenter trial on highly cross-linked polyethylene inserts in patients undergoing total hip arthroplasty, 118 patients consented to 10-year follow-up. Medium-term follow-up results showed low wear at 5 and 7 years after surgery. The current study focuses on long-term data at 10 years. Patients were followed up by either phone or office visit to collect long-term clinical data including Harris hip score and adverse events. There were 2 deaths and 2 revisions, 1 at 6.5 years for pelvic cyst and severe pain and another at 8.0 years for recurrent dislocation. For wear analysis, suitable radiographic images for 48 patients (52 hips) at minimum 10-year follow-up were available. Mean age of the cohort was 62.5 years (62% female). Femoral head penetration was measured using Martell's method from the radiographic images between the 6-week and the subsequent follow-ups. Polyethylene wear rate was calculated from the penetration data. Descriptive statistics were performed. There was no evidence of significant oxidation or locking mechanism failure. Mean Harris hip score of the group was 94.3. No osteolysis was noted by an independent radiographic reviewer. The slope of the bestfit regression line to the femoral head penetration data, which represents the overall linear wear rate of the polyethylene, was 0.014 mm/y. The wear rate was significantly below the 0.100 mm/y critical threshold for development of osteolysis. Volumetric wear rate was calculated to be 11.6 mm3/y. The second-generation highly cross-linked polyethylene acetabular inserts had low wear in the long term (10.3 years) with no incidence of osteolysis. [Orthopedics. 2018; 41(4):e529-e533.].

Once Annealed Highly Cross-Linked Polyethylene Exhibits Low Wear at 9 to 15 Years.

A once annealed highly cross-linked polyethylene (HXLPE) was introduced in 1998. Concerns regarding its long-term performance and oxidative resistance exist because of the presence of retained free radicals. The authors studied 48 patients with 50 hip implants having an average age of 62 years. They were followed for 9 to 15 years. The purpose of this study was to determine linear wear rate and the incidence of osteolysis and/or mechanical failure. At an average follow-up of 12.2 years, the annual linear wear rate was 0.018 mm (SD, 0.024 mm). No mechanical failures or osteolysis have been found to date. The clinical performance of this HXLPE continues to meet expectations despite the presence of free radicals. [Orthopedics. 2016; 39(3):e565-e571.].

Second-Generation Annealed Highly Crosslinked Polyethylene has Low Wear at Mean Seven Year Follow-up.

First-generation once annealed highly crosslinked polyethylene (HXLPE) has perfomed well for over 10 years but the presence of free radicals remain a concern. A second-generation sequentially annealed HXLPE was developed to further reduce wear, maintain mechanical strength, and enhance oxidative resistance.

Wear analysis of first-generation highly cross-linked polyethylene in primary total hip arthroplasty: an average 9-year follow-up.

This is a retrospective review of 46 primary total hip arthroplasties using a first-generation highly cross-linked ultrahigh-molecular-weight polyethylene liner (Crossfire; Stryker, Mahwah, NJ) with an average of 9-year follow-up. The purpose of this study was to measure linear penetration rate of first-generation polyethylene to determine if it maintains its wear resistance and fatigue strength over an extended period compared with conventional polyethylene. The mean (SD) total penetration was 0.339 (0.204) mm, and the mean (SD) penetration rate was 0.037 (0.022) mm/y. Our study demonstrated a 74% reduction in total penetration of highly cross-linked polyethylene when compared with historical controls using conventional polyethylene at an average of 9 years. Our results support the belief that highly cross-linked polyethylene does retain its wear resistance over an extended period.

Second-generation annealed highly cross-linked polyethylene exhibits low wear.

BACKGROUND: Compared with conventional polyethylene, first-generation highly cross-linked polyethylenes have low wear, but controversy exists regarding their reduced mechanical strength and/or retained free radicals. Second-generation highly cross-linked polyethylenes have been developed to reduce wear, maintain mechanical strength, and have oxidative resistance, but it is unclear whether they do so. QUESTIONS/PURPOSES: The primary objective of this study therefore was to determine if a second-generation annealed material has low linear wear at 5 years followup. Secondary objectives were to evaluate for overall survivorship, implant fixation, osteolysis, and effect of socket inclination on wear. METHODS: In a multicenter prospective study, we radiographically evaluated 155 patients (167 hips) at 3 years, 124 patients (132 hips) at 4 years, and 46 patients (51 hips) at 5 years. The linear head penetration rate was measured at 6 weeks, 1 year, and yearly through 5 years. RESULTS: The head penetration per year after the first year of bedding-in was 0.024 mm per year at 3 years, 0.020 mm per year at 4 years, and 0.008 mm per year at 5 years. The average wear rate over 5 years was 0.015 mm per year and represents a 58% improvement over a first-generation annealed highly cross-linked polyethylene. The Kaplan-Meier survivorship (revision for any reason) was 97.8%. We revised no hip for bearing surface failure and observed no osteolysis. Socket inclination did not affect linear wear. CONCLUSIONS: These data suggest the linear wear rate for a second-generation annealed highly cross-linked polyethylene is no greater than that for historic controls of first-generation highly cross-linked polyethylenes, and no untoward complications were encountered with this new material. LEVEL OF EVIDENCE: Level II, prognostic study. See the Guidelines for Authors for a complete description of levels of evidence.

Continued improved wear with an annealed highly cross-linked polyethylene.

BACKGROUND: Highly cross-linked polyethylene (HXLPE), created by disrupting the molecular structure of polyethylene, then through the application of heat, encourages creation of new cross-links in the process, resulting in a material with improved wear resistance. The impetuses for this new technology were the unsatisfactory wear properties and subsequent osteolysis of noncross-linked polyethylene. A 72% reduction in wear using highly cross-linked polyethylenes (HXLPE) compared with conventional polyethylene at 5 years was described previously. The longest term followup studies on HXLPE range from 2 to 6 years. QUESTIONS/PURPOSES: We therefore addressed the following questions: (1) Does the improvement in wear observed at the earlier followup continue to 7 to 10 years? (2) What is the incidence of osteolysis in this group of patients and in the control group? METHODS: We retrospectively reviewed 38 prospectively followed patients who had 42 hips with an annealed HXLPE who were followed a minimum of 7 years (average, 8.6 years; SD=1; range, 7-10.3 years). Wear and osteolysis were compared with those of a control group of 39 patients (40 hips) from a US Investigational Device Exemption (IDE) prospective, randomized study begun in 1996 with conventional polyethylene and followed for a minimum of 6 years (average, 7.5 years; SD=1.1; range, 6-10.2 years). Linear head penetration was measured from AP radiographs at early, 1-year, 5-year, and most recent followups. RESULTS: At the average followup, annual linear wear was 0.031 mm (SD=0.014) for the HXLPE and 0.141 mm (SD=0.080) for the control group, a 78% reduction. No mechanical failure of the polyethylene was noted in either group. Incidence of osteolysis was 50% in the control group (all lesions confined to proximal Gruen Zones 1 and 7) compared with no cases in the investigational group. CONCLUSIONS: We observed an improvement in wear and no mechanical failures with this annealed material. LEVEL OF EVIDENCE: Level III, therapeutic study. See Guidelines for Authors for a complete description of levels of evidence.

Five-year comparative study of highly cross-linked (crossfire) and traditional polyethylene.

The purpose of this study was to compare the linear penetration rates of a highly cross-linked ultra-high-molecular-weight polyethylene to traditional polyethylene. Twenty-seven highly cross-linked components were matched to 27 traditional polyethylene components with respect to age, sex, body mass index, and activity level. Each group was followed up for a mean of approximately 6 years. Linear penetration was measured using Martell's computerized technique (J Bone Joint Surg Am 1997;79:1635). The mean total penetration was 0.283 mm (SD, 0.253) for the highly cross-linked group. The mean total penetration for the traditional polyethylene was 0.696 mm (SD, 0.402). This difference was highly significant (P < or = .001). Our experience demonstrates a 59% reduction in total penetration of highly cross-linked polyethylene when compared with traditional polyethylene at a minimum of 5 years.

Metal-on-metal hip resurfacing radiographic evaluation techniques.

Hip resurfacing devices require a new radiographic evaluation technique owing to femoral components with short or no stems. Fourteen US surgeons implanted 1148 metal-on-metal hip resurfacing (HR) devices in a US-FDA-IDE clinical trial, which began in 2001. In this multi-center, prospective study, 337 patients (mean age, 50.1 years) were enrolled as a study group of unilateral HR arthroplasties. Radiographs of 292 HR arthroplasties at a minimum 2-year follow-up (maximum 3 years) were reviewed. There were 10 patients with radiographic evidence of femoral component instability beyond 2 years, as evidenced by subsidence > or = 5mm. Of these, 7 did not have clinical symptoms associated with femoral component instability. In the study group, 24 revisions were reported, of which 8 were due to femoral neck fractures, 4 were due to acetabular component loosening, 11 were due to femoral component loosening, and 1 due to dislocation.

Crosslinked polyethylene compared to conventional polyethylene in total hip replacement: pre-clinical evaluation, in-vitro testing and prospective clinical follow-up study.

BACKGROUND: Polyethylene wear-induced osteolysis is a major cause of implant loosening in total hip arthroplasty. New crosslinked polyethylenes are presumed to give lower wear rates, but no long-term clinical results are available yet. PATIENTS AND METHODS: We compared basic material characteristics and MTS hip joint simulator wear rates of a crosslinked polyethylene (Duration) to those of conventional polyethylene. In a randomized double-blind 5-year clinical follow-up study, 133 hips (67 conventional, 66 Duration) in 127 patients were followed-up for an average of 5 (3-6) years. Wear rates were measured using a computer-based edge detection method. The radiographic appearances of wear-related phenomena were recorded. RESULTS: The Duration polyethylene showed a significantly lower in-vitro wear rate in the simulator study (mean 22 (SD 2.3) vs. 40 (SD 1.5) mm3/106 cycles). Also, the in-vivo wear was lower for Duration (mean 0.083 (SD 0.056) mm/year) than for conventional polyethylene (mean 0.123 (SD 0.082) mm/year). All radiographic signs of osteolysis were less frequent in the Duration group. INTERPRETATION: Our study has given a substantial body of evidence--from lower wear rates, less frequent signs of osteolysis, and higher survival rates after a mean follow-up of 5 years--that Duration provides better clinical outcomes than conventional polyethylene.

Five-year experience with Crossfire highly cross-linked polyethylene.

UNLABELLED: Our purpose was to compare the clinical wear performance of highly cross-linked (Crossfire) polyethylene with conventional (N2Vac) polyethylene total hip bearings. We implanted 56 hips (47 patients) with Crossfire acetabular bearing liners and compared their wear performance with 53 conventional polyethylene inserts. Wear and clinical data were collected retrospectively at a minimum 4-year and average 5-year followup. The linear femoral head penetration rate measured from plain radiographs was 0.055 mm/year +/- 0.022 mm/year for the Crossfire polyethylene and 0.138 mm/year +/- 0.073 mm/year for the control, a reduction of 60% for the Crossfire components. Calculated annual wear was 0.036 mm/year for the Crossfire components and 0.131 mm/year for the controls, a reduction of 72%. Radiographic review at most recent followup showed a reduction in erosive osteolytic lesions of the proximal femur for the Crossfire components compared with controls, also suggesting a reduction in debris release for the Crossfire components. Complications leading to revision were not seen in the Crossfire or control groups. These clinical findings suggest that this particular highly cross-linked polyethylene can provide wear reduction and an alternate bearing surface for active patient populations. LEVEL OF EVIDENCE: Therapeutic study, Level III (retrospective comparative study). See the Guidelines for Authors for a complete description of levels of evidence.