Arterial catheter outcomes in intensive care: An analysis of 1117 patients.

BACKGROUND: Access to arterial circulation through arterial catheters (ACs) is crucial for monitoring and decision-making in intensive care units (ICU) but carries the risk of complications including bloodstream infection (BSI). METHODS: We conducted a secondary analysis of data from four randomised controlled trials in Australian ICUs, investigating the efficacy of different AC interventions. De-identified data were combined into a single dataset, and per-patient outcomes analysed. The primary outcome was AC-BSI, defined as laboratory confirmed bloodstream infection (LCBI) type 1 or 2, with a concurrent local infection. All-cause AC failure was defined as any unplanned removal. AC infection and failure were reported as rates per 1000 catheter days and hours. RESULTS: Data from 1117 adult patients were analysed. Mean age was 58.8 years (±16.6); and 41% (n = 462) were male. Median AC dwell time was 110 h (IQR 28.3-168.0). There was one case (<0.1%; 0.18/1000 catheter days [95% CI 0.03-1.29]) of AC-BSI, and 14 cases of LCBI (1%; 13 LCBI-1 and 1 LCBI-2; 2.54/1000 catheter days [95% CI 1.51-4.30]). LCBI were most commonly Enterococcus faecalis; Escherichia coli and Klebsiella pneumoniae. There were four cases of local infection (<1%; 0.73/1000 catheter days [95% CI 0.27-1.94]). Overall AC failure rate was 13% (n = 146) or 26.53/1000 catheter days (95% CI 22.56-31.20). CONCLUSION: This study identified a relatively low incidence of complications. This is likely reflective of poor monitoring of ACs in intensive care. Better surveillance and a rigorous prospective evaluation of AC outcomes is required to understand the true risk ACs pose to critically ill patients.

Cochlear reimplantation outcomes over 20 years: Expertise in reimplantation surgery and auditory-speech rehabilitation.

OBJECTIVES: The aim of this study was to present an institution's experience with cochlear reimplantation (CRI), to assess surgical challenges and post-operative outcomes and to increase the success rate of CRI. STUDY DESIGN: Retrospective single-institution study. SETTING: Tertiary medical center. METHODS: We retrospectively evaluated data from 76 reimplantation cases treated in a tertiary center between 2001 and 2022. Clinical features including etiology of hearing loss, type of failure, surgical issues, and auditory speech performance were analyzed. Categorical Auditory Performance (CAP) and Speech Intelligibility Rating (SIR) scores were used to evaluate pre- and post-CRI outcomes. RESULTS: The CRI population comprises of 7 patients from our institute,69 referred patients from other centers. Device failure was the most common reason (68/76, 89.5 %) for CRI; in addition, there were 7 medical failures and 1 had both soft device failure. Medical failures included flap rupture and device extrusion, magnet migration, auditory neuropathy, leukoencephalopathy, foreign-body residue and meningitis. In 21/76 patients, the electrode technology was upgraded. The mean time to failure was 0.58-13 years, with a mean of 4.97 years. The mean (± SD) CAP and SIR scores before and after CRI were 5.2 ± 1.2 versus 5.5 ± 1.1 and 3.4 ± 1.1 versus 3.5 ± 1.1, respectively. Performance was poor in six patients with severe cochlear malformation, auditory nerve dysplasia, leukoencephalopathy, and epilepsy. CONCLUSION: CRI surgery is a challenging but relatively safe procedure, and most reimplanted patients experience favorable postoperative outcomes. Medical complications and intracochlear damage are the main causes of poor postoperative results. Therefore, adequate preoperative preparation and atraumatic CRI should be carried out for optimal results.

Adverse events related to suction electrocautery devices in adenotonsillectomy: Analysis of the MAUDE database.

OBJECTIVE: To identify adverse events (AEs) related to suction electrocautery use during adenotonsillectomy. METHODS: The US Food and Drug Administration (FDA) Manufacturer and User Facility Device Experience (MAUDE) database was searched using the terms "suction cautery," "suction electrocautery," "suction Bovie," and "suction coagulator" from January 2014 to December 2023. RESULTS: 165 AE reports were gathered from the MAUDE database medical device reports (MDRs). 36 met inclusion criteria. Patient injuries were found in 22 (61.1 %) reports and device malfunction events were found in 14 (38.9 %) reports. All patient injuries were thermal burns (N = 22, 100 %). Location of burn injuries included the lip (N = 6, 27.3 %), oral commissure (N = 5, 22.7 %), and tongue (N = 4, 18.2 %). The most common cause of an AE was inadequate device insulation (N = 7, 19.4 %). CONCLUSION: The suction electrocautery apparatus may malfunction and cause patient burn injuries. Device failures mainly result from inadequate device insulation, coagulation problems, and detachment of device components. Surgeons must be aware of these potential complications and counsel parents and patients regarding AEs.

Revision Cochlear Implantation With Device Manufacturer Conversion: Surgical Outcomes and Speech Perception Performance.

OBJECTIVE: Describe the clinical profile of revision cochlear implantation (RCI) cases involving device manufacturer conversion (RCImc+), compare them to cases without manufacturer conversion (RCImc-), and classify the reasons for manufacturer conversion (MC). STUDY DESIGN: Retrospective case review. SETTING: Tertiary academic center. METHODS: Data on demographics, RCI indications, medical background, surgical details, and the reasons for MC were collected for all RCIs from 1989 to 2020. Post-RCI speech perception performance was categorized as unchanged, improved, or declined, according to clinically based criteria. RESULTS: Of 185 RCIs, 39 (21%) involved MC, mostly in pediatric patients (67%). The leading RCImc+ indications were device-related (59%) and medical (31%) failures. Initial implant manufacturers were Advanced Bionics (49%), Cochlear (25.5%), or Medel (25.5%). Most MC reasons were patient-driven (64%) versus CI team recommendations (36%). The RCImc+ group demonstrated a 3-fold higher rate of medical indications than RCImc- (31% vs 11.5%, P = .007). The time interval from symptom onset to RCI was longer in RCImc+ (43 vs 20.3 months, P = .001), and the rate of multiple revisions in the same ear was higher (25.6% vs 8.2%, P = .009). Complete reinsertion rates were high in both RCImc+ and RCImc- (94.8% vs 94.5%, P = 1) without any complications. Speech perception improved or remained unchanged in most (84%) cases, with no significant difference between the groups (P = .183). CONCLUSION: This retrospective study showed that RCI involving MC is safe and beneficial. Although RCImc+ patients exhibited distinct clinical characteristics, MC did not impact surgical or speech perception outcomes. This provides evidence-based data to support informed decision-making by CI teams and patients.

Single suture-mediated closure system after transfemoral transcatheter aortic valve implantation: A single-center real-world experience.

BACKGROUND: Despite the use of two crossed Perclose ProGlide™ (Abbott Vascular Devices) is the most widespread technique to close the main arterial access in transfemoral transcatheter aortic valve implantation (TF-TAVI), the safest and most effective strategy still remains much debated. AIMS: The aim of the present study was to evaluate the performance of a single Perclose ProGlide suture-mediated closure device to obtain femoral hemostasis after sheathless implantation of self-expanding transcatheter heart valves through their 14 F-equivalent fix delivery systems. METHODS: This prospective observational study included 439 patients undergoing TF-TAVI at the "Montevergine" Clinic of Mercogliano, Italy. All patients underwent hemostasis of the large-bore access using a single Perclose ProGlide with preclose technique, after sheathless implantation of self-expanding transcatheter heart valves through 14 F-equivalent fix delivery systems. A multidetector computed tomography analysis of size, tortuosity, atherosclerotic, and calcification burdens of the ilio-femoral access route was made by a dedicated corelab. Vascular complications (VCs), percutaneous closure device (PCD) failure, and bleedings were adjudicated by a clinical events committee. RESULTS: A total of 81 different VCs were observed in 60 patients (13.7%); among these, 41 (5% of patients) were categorized as major. PCD failure occurred in 14 patients (3.2%). At the logistic regression analysis, no predictors of PCD failure have been identified. CONCLUSION: This registry suggests that the use of a single suture-mediated closure device could be considered a safe and efficient technique to achieve access site hemostasis in patients undergoing TF-TAVI through 14 F-equivalent fix delivery systems.

Postoperative complications in cochlear implant surgery and their possible risk factors.

OBJECTIVE: This is a retrospective analysis of the major and minor complications of cochlear implants, as well as the Risk Factors (RF) involved. METHODS: We analyzed the medical records of patients submitted to cochlear implants at public University from 2006 to July 2019, and list here the major and minor complications found, and their risk factors. RESULTS: There were 193 ears, 100 (51.3%) from females and 93 (48.2%) from males, with a mean age of 23.63 years. In 54 of them (28%), there were alterations seen in the Temporal Bone CT scan, and 44 (22.8%) in the brain MRI. There were 158 (81.9%) insertions performed; 127 (65.8%) of them through the round window. There were 78 complications: 19 (9.8%) major and 56 (29%) minor complications. Among the major complications, there were 3 (1.6%) Surgical Site infections (SS); 5 (2.6%) hematomas/seromas; 5 (2.6%) electrode extrusion; 5 (2.6%) device faults; 1 (0.5%) wrong path. Among the minor complications, there were 6 (3.1%) Acute Otitis Media (AOM); 9 (4.7%) SS infections; 4 (2.1%) facial paresis; 17 (8.8%) vertigos; 9 (4.7%) with tinnitus. The most important RF was age. Patients younger than 2.5 years had more major complications: SS infection (p = 0.018) and electrode extrusion (p = 0.017). There was a higher rate of vertigo in adults (p = 0.003), and it was more often associated with comorbidities (p = 0.008). The insertion route, the presence of changes in CT and MRI and the CI brand used did not impact the number of complications. CONCLUSION: Among the minor complications, those involving the vestibular system were the most common, especially in adults with comorbidities. Regarding major complications, there was an emphasis on SS infections, hematomas, seromas, electrode extrusion, especially in children under two years of age. There were implanted device faults (2.6%), with none of the brands evaluated standing out.

Paediatric percutaneous bone anchored hearing aid implant failures: Comparing the experience of a tertiary centre with a systematic review of the literature and meta-analysis.

BACKGROUND: Despite the proven audiological benefits of Percutaneous Bone Anchored Hearing Aids (BAHAs) in paediatric patients with conductive or mixed hearing loss, their adoption has been limited due to concerns over implant failure and associated complications. This paper conducts a systematic review and meta-analysis to assess the prevalence of implant failure in paediatric populations, combined with a case series from our tertiary referral centre. METHODS: A comprehensive literature search identified 562 articles, from which 34 were included in the review, covering 1599 implants in 1285 patients. Our retrospective case series included consecutive patients from our tertiary referral centre who underwent percutaneous BAHA implantation from 2003-2019. RESULTS: Meta-analysis revealed an overall implant failure rate of 11%, predominantly attributed to traumatic extrusion. Our retrospective case series comprised 104 implantations in 76 patients, with a 4.8% failure rate. DISCUSSION: Factors contributing to the lower-than-expected failure rates in the case series likely included consistent use of 4 mm fixtures from a single manufacturer and older age at implantation. The study underscores the need for standardised reporting formats in bone conduction implants research, given the systematic review's limitations in study design heterogeneity, especially with the expected rise in the adoption of novel active devices.

Uncommon Carotid Artery Stenting Complications: A Series by Images.

AIMS: To describe through emblematic images rare but clinically relevant carotid artery stenting complications that occurred at two high-volume centres for carotid artery stenting (CAS). BACKGROUND: CAS is an alternative to carotid endarterectomy (CEA) for the treatment of carotid artery stenosis in patients judged to be at high risk for CEA. CAS complications range between 1 and 9% and are higher in older patients complaining of neurological symptoms at the time of presentation. Besides periprocedural or early-after-procedure stroke, which remains the true Achilles' heel of CAS, other dramatic complications might compromise the clinical outcomes of this procedure. METHODS: Five infrequent complications, out of more than 1000 CAS performed in the years 2016-2021, have been described. RESULTS: Among CAS complications, acute carotid stent thrombosis, rescue retrieval of a disconnected distal cerebral embolic protection device, plaque prolapse after carotid stenting, cerebral hyperperfusion syndrome (CHS), and radial artery long sheath entrapment requiring surgical intervention were found to account for 0.3% of the total number of procedures performed by operators with high CAS volume. CONCLUSIONS: Unusual CAS complications may infrequently occur, even in hands of expert operators. To know how to deal with such complications might help interventionalists to improve CAS performance.

Tough balloon with a vulnerable shaft: an unexpected device failure.

A 73-year-old man, with a previous long drug-eluting stent (DES) from the mid-portion of the left main stem artery (LMS) to proximal left anterior descending artery (LAD), underwent elective coronary angiogram due to worsening anginal symptoms and an abnormal myocardial perfusion single photon emission computed tomography (SPECT) showing ischemia in the LAD territory.

Adverse events associated with the gold probe and the injection gold probe devices used for endoscopic hemostasis: A MAUDE database analysis.

BACKGROUND: Gastrointestinal (GI) bleeding accounts for over half a million admissions annually and is the most common GI diagnosis requiring hospitalization in the United States. Bipolar electrocoagulation devices are used for the management of gastrointestinal bleeding. There is no data on device-related adverse events for gold probe (GP) and injection gold probe (IGP). AIM: To analyze this using the Food and Drug Administration (FDA's) Manufacturer and User Facility Device Experience (MAUDE) database from 2013 to 2023. METHODS: We examined post-marketing surveillance data on GP and IGP from the FDA MAUDE database to report devicerelated and patient-related adverse events between 2013-2023. The MAUDE database is a publicly available resource providing over 4 million records relating to medical device safety. Statistical analyses were performed using IBM SPSS Statistics V.27.0 (IBM Corp., Armonk, NY, United States). RESULTS: Our search elicited 140 reports for GP and 202 reports for IGP, respectively, during the study period from January 2013 to August 2023. Malfunctions reportedly occurred in 130 cases for GP, and actual patient injury or event occurred in 10 patients. A total of 149 patients (74%) reported with Injection GP events suffered no significant consequences due to the device failure, but 53 patients (26%) were affected by an event. CONCLUSION: GP and IGP are critical in managing gastrointestinal bleeding. This study of the FDA MAUDE database revealed the type, number, and trends of reported device-related adverse events. The endoscopist and support staff must be aware of these device-related events and be equipped to manage them if they occur.

Changes in Revision Cochlear Implantation and Device Failure Profiles.

OBJECTIVES: As cochlear implantation (CI) experiences rapid innovations and its indications expand, the characteristics of revision CI (RCI) are evolving. This study investigated changes in the RCI profile and explored their clinical implications. METHODS: A retrospective chart review was conducted of all CIs performed at a tertiary medical institution between October 2001 and January 2023. The rates of and reasons for RCI were evaluated in relation to the manufacturer and device model. Kaplan-Meier analysis was employed to examine cumulative and device survival curves. Cumulative and device survival rates were additionally analyzed based on age group, period of primary CI, and manufacturer. A Cox proportional hazards model was employed to evaluate the association between RCI and the device manufacturer. RESULTS: Among 1,430 CIs, 73 (5.1%) required RCI. The predominant reason for RCI was device failure (40 of 73 RCIs [54.8%]), with an overall device failure rate of 2.8%. This was followed by flap-associated problems and migration (nine of 73 RCIs each [12.3%]). Flap retention issues emerged as a new cause in three cases (two involving the CI 632 and one involving the SYNCHRONY 2 implant), and six instances of electrode tip fold-over arose (four for the CI 600 series and two for the CI 500 series). The overall 10-year cumulative and device survival rates were 93.4% and 95.8%, respectively. After excluding models with recall issues, significant differences in cumulative (P =0.010) and device (P =0.001) survival rates were observed across manufacturers. CONCLUSION: While the overall CI survival rate is stable, device failure persists as the predominant reason for RCI. Moreover, the types of complications leading to revision (including issues with flap retention and electrode tip fold-over) have shifted, particularly for newer implant models. Given the clinical importance of device failure and subsequent reoperation, clinicians should remain informed about and responsive to these trends.

Upper airway stimulation device failure: A 7-year single center experience.

OBJECTIVE: To identify and report a single center experience with upper airway stimulator device-related failures. STUDY DESIGN: Retrospective case series. SETTING: Single tertiary academic center. METHODS: Retrospective data on 352 patients who underwent UAS surgery with an Inspire device from 2016 to 2023 was collected, including demographics, comorbidities, and nature of device failure requiring revision surgery. RESULTS: Out of the 348 patients included in our analysis, 16 (4.6 %) required revision due to device failure, with an average interval of 772 days (∼2 years) between initial implant and revision. Most failures were attributed to respiratory sensing lead damage (n = 11, 68.8 %), resulting in high system impedance and subsequent device malfunction. Lead fracture causes varied, including idiopathic occurrences and potential trauma. Lead migration was noted in one case (6.3 %), where the hypoglossal electrode detached from the nerve. Two patients (12.3 %) required implantable pulse generator (IPG) replacement, one after experiencing trauma and the other due to unclear source of malfunction. One patient (6.3 %) required complete system replacement following high lead impedance and absent tongue motion. The last patient required replacement of both the IPG and respiratory lead after experiencing high lead impedance (6.3 %). CONCLUSION: Respiratory sensing lead fracture emerged as the leading cause of device failure in this cohort, underscoring the need to address this under-reported issue, potentially linked to the time lapse after device implantation.

Microdebrider complications in sinus surgery: Analysis of the openFDA database.

Objective: Functional endoscopic sinus surgery is a commonly performed otolaryngologic procedure that often uses the microdebrider device for tissue removal. Given the ubiquitous nature of the instrument, we sought to better define the patterns of device failure using the postmarket surveillance openFDA database. Methods: The openFDA database was queried for all microdebrider-related adverse events from January 1, 2000 to November 1, 2020. Descriptive information on the nature of device failure and any associated patient injury was compiled. Reports not directly related to device failure were excluded from the analysis. Results: A total of 641 events were included in the analysis. The most common device failure was overheating (n = 348, 54.3%), followed by material separation (n = 173, 27%), and inconsistent device activation (n = 52, 8.1%). Of the reported events, the vast majority did not result in patient harm (n = 579, 90.3%). On review of the remaining cases, only 24 events (3.7%) resulted in true harm to the patient, defined as a temporary or permanent injury or >30 min of additional anesthesia time. Of these cases, the need to reschedule surgical cases (n = 5, 0.8%), retained foreign body (n = 5, 0.8%), and thermal tissue injury (n = 3,0.5%) were the most common. Five patients suffered an injury due to surgeon error unrelated to device malfunction (n = 5, 0.8%). Conclusions: Microdebrider device failures are extremely rare. When they do occur, less than 10% result in patient harm. In cases of patient harm related to microdebrider failure, preoperative testing of the device before use could prevent many of the reported malfunctions.

Revision cochlear implant surgery in adults and children is a safe and effective technique: a retrospective single faculty study in Okayama University Hospital, Japan.

BACKGROUND: Cochlear implant (CI) surgery is a safe surgical technique, although some patients require revision CI surgery. AIMS/OBJECTIVES: This study investigated the cause and underlying reason of revision CI surgery as well as hearing outcomes in a single institution. PATIENTS AND METHODS: This retrospective study evaluated patients who underwent CI surgery between April 2006 to March 2022 (n = 351). Sex, aetiology of hearing loss (HL), age and period from initial CI surgery to reimplantation, cause of revision, and related factors were examined. RESULTS: Twelve patients (8 males, 4 females) received CI reimplantation. The revision surgery rate was 2.59% (3.15% children, 1.69% adults); the period from initial surgery to reoperation was 8.60 ± 6.56 years for 9 children with congenital HL and 15.27 ± 5.72 years for 3 adults with progressive HL. Device failure was the most common cause (n = 8), followed by infections (n = 2), advanced facial irritation symptoms (n = 1), and electrode slip-out (n = 1). Mean preoperative and postoperative CI thresholds were 44.0 ± 9.46 dBnHL and 39.19 ± 8.89 dBnHL (p < .068), respectively. CONCLUSION AND SIGNIFICANCE: Caregiver education, surgical technique advances, flap design, and extensive antibiotic use may decrease the revision surgery rate. The lack of post-revision deterioration of the hearing threshold contributed to well-being in patients with CI.

The Use of Piezoelectric Osteotomy Instruments During a Rhinoplasty Operation in a Patient With a Cochlear Implant.

A cochlear implant is a life-changing operation that significantly improves the lives of patients. Those with cochlear implants require ongoing measures to ensure the longevity of both their devices and their well-being. One concerning issue is the potential device failure due to the use of surgical instruments during procedures. In this report, we present a successful case of utilizing an ultrasonic osteotome for a primary septorhinoplasty in a 22-year-old patient who had undergone cochlear implant surgery 17 years prior. To our knowledge, this is the first recorded instance of applying a piezoelectric tool on a patient with a cochlear implant. Our findings support the safety of using a piezo osteotome in cochlear implant recipients.

Strategy for Totally Percutaneous Management of Vascular Injury in Combined Transfemoral Transcatheter Aortic Valve Replacement and Endovascular Aortic Aneurysm Repair Procedures.

Minimally invasive treatment of severe aortic stenosis by transcatheter aortic valve replacement (TAVR) and infrarenal abdominal aortic aneurysm by endovascular aortic aneurysm repair (EVAR) requires large-bore vascular access. These percutaneous transfemoral interventions may be performed as a combined procedure, however, vascular injury may necessitate surgical vascular repair. We implemented a strategy designed to enable percutaneous vascular repair, with stent-graft implantation, if necessary, after these combined procedures. We identified all combined percutaneous TAVR and EVAR procedures which were performed at our institution. Patient and procedural characteristics and clinical outcomes were analyzed. Six consecutive patients underwent total percutaneous combined TAVR and EVAR procedures. In all cases, TAVR was performed first and was followed by EVAR. Both common femoral arteries served as primary access sites for delivery of the implanted devices and hemostasis was achieved by deployment of vascular closure devices. Secondary access sites included the right brachial artery in all patients and superficial femoral arteries in 50% of the patients. In all cases an "0.014" 300-cm length "safety" wire was delivered to the common femoral artery or descending aorta by way of a secondary access site to facilitate stent graft delivery. Successful device implantation was achieved in all cases. Vascular closure device failure occurred in 2 patients and was treated by stent graft implantation by way of the brachial and superficial femoral arteries, without need for surgical vascular repair. A strategy designed to facilitate percutaneous vascular repair after combined EVAR and TAVR procedures may enable a truly minimally invasive procedure.

Endovascular Graft Suprarenal Bare Metal Stent Separation After Endovascular Aneurysm Repair: Case Reports and Literature Review.

Objective: Suprarenal bare metal stent separation is a rare complication after endovascular aneurysm repair. In this report, two new cases of this type of device failure are presented and the literature is reviewed to identify similar cases and evaluate associated clinical characteristics. Methods: A literature search was conducted in March 2022 using PubMed, Embase, and The Cochrane Library, with MeSH terms including aortic aneurysm, stents, and device failure. Two authors independently selected studies eligible for inclusion. Results: Twelve patients with endovascular graft suprarenal bare metal stent separation were identified. Endovascular aneurysm repair (EVAR) devices were implanted between May 1996 and November 2017. Suprarenal bare metal stent separation was detected after a median duration of five years post-operatively. Conclusion: Endovascular graft suprarenal bare metal stent separation demands a high level of awareness. A better understanding of the involved failure mechanisms and associated risk factors is required to further optimise EVAR follow up protocols.

The experience of device failure after cochlear implantation.

BACKGROUND: The present study describes the treatment of patients at a tertiary institution who experienced device failure after Cochlear Implantation (CI), as well as identifying prodromic symptoms that could assist in the timely identification and management of device failure. STUDY DESIGN: Retrospective database review (January 2000-May 2017). SETTING: Single tertiary hospital. METHODS: Factors recorded included the etiology of hearing loss; age at first and revision CI surgeries; surgical information, including operation time and approach; electrical outcomes after implantation; device implanted; symptoms of device failure; history of head trauma; and audiologic outcomes as determined by categories of auditory performance (CAP). RESULTS: From January 2000 to May 2017, 1431 CIs were performed, with 27 (1.9%) undergoing revision surgeries due to device failure. The most common etiology of hearing loss was idiopathic (12/27), followed by cochlear hypoplasia (5/27). Mean age at initial CI was 11.8 (1-72) years, with 21 being pre-lingual and 6 being post-lingual. Of the total devices initially implanted, 80.5% were from Cochlear, 15.9% from MED-EL, and 3.5% from Advanced Bionics. The failure rates of these devices were 1.3%, 3.1%, and 10.0%, respectively. The most suggestive symptom of device failure was intermittent loss of signal. Mean CAP scores were 5.17 before reimplantation and 5.54 and 5.81 at 1- and 3-years, respectively, after reimplantation. CONCLUSION: The most suggestive symptom preceding device failure was intermittent loss of signal. Patients who present with this symptom should undergo electrical examination for suspected device failure. Audiologic outcomes showed continuous development despite revision surgeries.

Predicting medical device failure: a promise to reduce healthcare facilities cost through smart healthcare management.

Background: The advancement of biomedical research generates myriad healthcare-relevant data, including medical records and medical device maintenance information. The COVID-19 pandemic significantly affects the global mortality rate, creating an enormous demand for medical devices. As information technology has advanced, the concept of intelligent healthcare has steadily gained prominence. Smart healthcare utilises a new generation of information technologies, such as the Internet of Things (loT), big data, cloud computing, and artificial intelligence, to completely transform the traditional medical system. With the intention of presenting the concept of smart healthcare, a predictive model is proposed to predict medical device failure for intelligent management of healthcare services. Methods: Present healthcare device management can be improved by proposing a predictive machine learning model that prognosticates the tendency of medical device failures toward smart healthcare. The predictive model is developed based on 8,294 critical medical devices from 44 different types of equipment extracted from 15 healthcare facilities in Malaysia. The model classifies the device into three classes; (i) class 1, where the device is unlikely to fail within the first 3 years of purchase, (ii) class 2, where the device is likely to fail within 3 years from purchase date, and (iii) class 3 where the device is likely to fail more than 3 years after purchase. The goal is to establish a precise maintenance schedule and reduce maintenance and resource costs based on the time to the first failure event. A machine learning and deep learning technique were compared, and the best robust model for smart healthcare was proposed. Results: This study compares five algorithms in machine learning and three optimizers in deep learning techniques. The best optimized predictive model is based on ensemble classifier and SGDM optimizer, respectively. An ensemble classifier model produces 77.90%, 87.60%, and 75.39% for accuracy, specificity, and precision compared to 70.30%, 83.71%, and 67.15% for deep learning models. The ensemble classifier model improves to 79.50%, 88.36%, and 77.43% for accuracy, specificity, and precision after significant features are identified. The result concludes although machine learning has better accuracy than deep learning, more training time is required, which is 11.49 min instead of 1 min 5 s when deep learning is applied. The model accuracy shall be improved by introducing unstructured data from maintenance notes and is considered the author's future work because dealing with text data is time-consuming. The proposed model has proven to improve the devices' maintenance strategy with a Malaysian Ringgit (MYR) cost reduction of approximately MYR 326,330.88 per year. Therefore, the maintenance cost would drastically decrease if this smart predictive model is included in the healthcare management system.