استكشف المجموعة الواسعة من العناوين المتاحة.

Medication errors account for a large proportion of all medical errors. In most homes, patients take a variety of medications for a long period. However, medication errors frequently occur because patients often throw away the containers of their medications. We proposed a deep learning-based system for reducing medication errors by accurately identifying prescription pills. Given the pill images, our system located the pills in the respective pill databases in South Korea and the United States. We organized the system into a pill recognition step and pill retrieval step, and we applied deep learning models to train not only images of the pill but also imprinted characters. In the pill recognition step, there are 3 modules that recognize the 3 features of pills and their imprints separately and correct the recognized imprint to fit the actual data. We adopted image classification and text detection models for the feature and imprint recognition modules, respectively. In the imprint correction module, we introduced a language model for the first time in the pill identification system and proposed a novel coordinate encoding technique for effective correction in the language model. We identified pills using similarity scores of pill characteristics with those in the database. We collected the open pill database from South Korea and the United States in May 2022. We used a total of 24,404 pill images in our experiments. The experimental results show that the predicted top-1 candidates achieve accuracy levels of 85.6% (South Korea) and 74.5% (United States) for the types of pills not trained on 2 different databases (South Korea and the United States). Furthermore, the predicted top-1 candidate accuracy of our system was 78% with consumer-granted images, which was achieved by training only 1 image per pill. The results demonstrate that our system could identify and retrieve new pills without additional model updates. Finally, we confirmed through an ablation study that the language model that we emphasized significantly improves the pill identification ability of the system. Our study proposes the possibility of reducing medical errors by showing that the introduction of artificial intelligence can identify numerous pills with high precision in real time. Our study suggests that the proposed system can reduce patients' misuse of medications and help medical staff focus on higher-level tasks by simplifying time-consuming lower-level tasks such as pill identification.

Various single‐pill combinations (SPCs) have been introduced to improve drug compliance and clinical efficacy. However, there is a lack of real‐world evidence regarding the effectiveness of these SPCs for hypertension. This study evaluated the real‐world clinical efficacy and safety of amlodipine/losartan‐based SPC therapies in patients with hypertension in a real‐world setting. A total of 15 538 patients treated with amlodipine/losartan‐based SPCs [amlodipine + losartan (AL), amlodipine + losartan + rosuvastatin (ALR), and amlodipine + losartan + chlorthalidone (ALC)] were selected from the database of three tertiary hospitals in Korea. The efficacy endpoints were target blood pressure (BP) and low‐density lipoprotein cholesterol (LDL‐C) achievement rates. Safety was evaluated based on laboratory parameters. Drug adherence was defined as the proportion of medication days covered (PDC). The target BP attainment rate was above 90% and was similar among the three groups. Although many patients in the AL and ALC groups took statins, the target LDL‐C attainment rate was significantly higher in the ALR group than in the AL and ALC groups. Safety endpoints were not significantly different among the groups, except serum uric acid level and incidence rate of new‐onset hyperuricemia, which were significantly lower in the AL and ALR groups than in the ALC group. The PDC was > 90% in all groups. In the real‐world hypertensive patients, amlodipine/losartan‐based SPC therapy demonstrated good target BP achievement rates. Especially, rosuvastatin‐combination SPC showed better target LDL‐C goal achievement rate compared to the other SPCs. All three amlodipine/losartan‐based SPC had excellent drug adherence.

IntroductionEthambutol, an antibiotic used to treat TB, is mostly well-tolerated and has a good safety profile. However, there is a rare but significant risk of vision damage, which in severe cases can be permanent. Stopping treatment promptly at symptom onset reduces risk. In 2017, following two incidents in the United Kingdom of severe visual loss in adults due to delayed intervention, The Royal College of Ophthalmologists issued a statement recommending baseline visual assessment prior to treatment start, with patients advised to report changes.1 Despite this, there seems a lack of clear guidance.This research project aimed to undertake a service evaluation exploring how TB teams in England monitor vision in ethambutol-treated patients, to highlight good practice and reveal potential service improvements.MethodsHaving gained ethical approval to survey NHS staff, online questionnaires were sent to 106 lead TB specialist nurses in England, and 53 responses were analysed using descriptive and inferential tests.ResultsOf 53 TB nurse respondents, 34 (64%) led acute trust TB nursing teams, and 19 (36%) led community trust TB nursing teams. Due to questionnaire branching, not all questions were answered by respondents. Visual assessments were provided by 51 respondents (96%). There were minimal differences in overall vision assessment provision, although more testing was conducted by TB nurses than by other practitioners (n=37; 46%). Qualitative comments from some TB nurses suggested a lack of guidance and protocols for conducting vision assessments. The majority of 45 respondents (n=36; 80%), relied on patients to self-report vision changes after initial baseline tests. Although no visual assessment telehealth solutions are currently validated for clinical use, 7 (15%) of 46 teams responding used these.Abstract P138 Table 1Who conducted vision testingPractitioner who conducted testing Total responses Prescribing physiciann=13 (15%) Tuberculosis nursen=37 (46%) Ophthalmologistn=17 (21%) Community optometristn=4 (5%)Outpatient clinic nurse/Healthcare assistantn=10 (13%) Overall totaln=81 (100%)ConclusionsAs visual assessments were primarily conducted by TB nurses, staff training and protocols should be reviewed, and appropriate testing resources made available. Reliance on patients self-reporting vision changes highlighted importance of patient counselling at the start of treatment. Clinically validated telehealth monitoring resources are also needed, with future innovations anticipated in this emerging and growing technological field.ReferenceRoyal College of Ophthalmologists (RCOphth) (2017) RCOphth Statement on Ethambutol Toxicity. Available at: https://www.rcophth.ac.uk/2017/10/rcophth-statement-on-ethambutol-toxicity/

IntroductionMulti-drug resistant tuberculosis (MDR-TB) remains a challenge despite shorter treatment regimes, with significant adverse effects that can impact psychological and physical aspects of health, potentially affecting treatment outcomes.MethodsA retrospective cohort study was completed at an East London teaching hospital including patients with confirmed MDR-TB or receiving MDR-TB drugs between 2010 and 2023.ResultsWe evaluated 32 patients who were started on MDR-TB treatment. Thirty patients had MDR-TB and two patients were on MDR treatment due to drug intolerance. Nineteen patients (59%) were male, median participant age was 39 years (IQR=20). Twenty-six patients out of 32 were migrants. Nineteen patients (59%) had pulmonary or pleural MDR-TB, eight patients (25%) had lymph node MDR-TB, with the remainder (16%) having MDR-TB in other sites including the renal system, skin, central nervous system, thyroid gland and/or joints. Twenty-eight patients were initiated on current WHO recommended treatment regimens for MDR-TB. Four patients were commenced on standard TB treatment regimes prior to being switched.Across our cohort, 24 patients (75%) developed an adverse effect to one or more drugs resulting in a treatment change. The most common drugs causing adverse effects resulting in a treatment change were linezolid (16/24), levofloxacin (5/24), pyrazinamide (10/24) and cycloserine (6/24).Linezolid was predominantly stopped (62%) due to symptoms suggestive of peripheral neuropathy. Other adverse events that resulted in treatment change included a metallic taste, glossitis, mood disturbance, anaemia, or serotonin syndrome.Levofloxacin was stopped in five patients due to joint pain and/or tendinopathies. Pyrazinamide was stopped in ten patients due to elevated urate levels and clinical symptoms of gout. Cycloserine was stopped in six patients due to low mood and suicidal ideations.ConclusionDespite newer more effective MDR-TB regimens, patients on MDR treatment continue to have significant side effects and should have access to shorter drug regimens.

Ocular tuberculosis (TB) is a rare extrapulmonary manifestation of TB, which can cause blindness. Diagnostic accuracy is a clinical challenge and implementing anti-tuberculous therapy (ATT) is not without risk. BTS clinical statement for the diagnosis and management of ocular TB1 encourages input from TB and ophthalmic specialists. We reviewed ocular response to ATT at our tertiary centre, aiming to improve patient information pre-treatment.Cases from our tertiary centre reported to Enhanced TB Surveillance System with ‘uveitis’, ‘ocular’ or ‘cryptic disseminated’ TB between 1st January 2016 to 31st January 2022 were identified. An additional search was performed using electronic patient records. Data on demographics, assessment, treatment and ocular response were collected.17 patients were included, 10 were male (58.5%) with mean age of 47 years. All patients had evidence of TB exposure and 82.5% had an interferon-gamma release assay (IGRA). All patients with an abnormal CXR (17.6%) went to have a CT. Of those who had an abnormal CT, all proceeded to diagnostic sampling (42.3%). Patients had eye changes suggestive of TB: choroidal granulomas (11.8%), mutton fat keratic precipitates (5.9%), serpiginous chorioretinopathy (5.9%) and peripheral occlusive retinal vasculitis (17.6%). The intended treatment for patients were between 9–12 months (70.6%). However, less than half (41.1%) completed treatment within recommended 365 days; some had ATT temporarily suspended owing to side effects.Of the 15 patients with confirmed ocular TB and available follow-up, six patients had evidence of systemic disease. One patient had microbiologically confirmed TB (6.7%) and five had imaging, histology and serology consistent with a diagnosis of TB (33.3%). At the end of treatment, over half of patients had quiescent eye disease (60.0%).In our centre, almost half our patients required further treatment for their ocular disease after completion of ATT, including patients who had evidence of TB beyond the eye. This is important information to set patient expectation at the beginning of ATT and a larger cohort is needed to confirm this outcome.Abstract P135 Table 1Patients confirmed to have Ocular TB with follow-up outcomes n = 15* Ocular TB with confirmed systemic TB IGRA/TST +veAbnormal CT imaging TB microbiology +ve Ocular TB with possible systemic TB IGRA/TST +veAbnormal CT imaging TB microbiology -ve Ocular TB alone IGRA/TST +veNormal CT imagingNo diagnostic sampling for microbiology Ocular TB with incomplete diagnostics IGRA/TST +veNo CT imaging or diagnostic sampling for microbiology Total no. of patients (% of cohort) 1 (6.7) 5 (33.3) 2 (13.3) 7 (46.7) No. of patients with quiescent eye disease following ATT (% of cohort) 0 (0) 3 (20.0) 0 (0.0) 6 (40.0) No. of patients with active eye disease following ATT (% of cohort ) 1 (6.7) 2 (13.3) 2 (13.3) 1 (6.7) *Two patients lost to follow-up were excludedReferenceKon OM, Beare N, Connell D, et al. BTS clinical statement for the diagnosis and management of ocular tuberculosis. BMJ Open Resp Res. 2022;9:e001225. doi:10.1136/bmjresp-2022-001225

IntroductionThe diagnosis of spinal tuberculosis (TB) is often delayed, which may affect treatment outcomes. NICE Guidelines suggest to treat spinal TB for 6 months however it does not take into account severity of disease (multi-level involvement or presence of paravertebral or psoas abscesses) at presentation.MethodsWe conducted a retrospective study of consecutive patients attending a spinal TB clinic at a large London Hospital. Information on demographics, clinical manifestation at diagnosis, radiology and treatment outcomes were recorded.ResultsSeventy-four cases with Spinal TB were identified during 2010–2022. Twenty-five cases had incomplete data. Out of 49 cases, 30(61%) were male and median age was 33 years (14 IQR). Out of 49 cases 42 (85%) were migrants and 7 cases were born in the UK.Forty-six patients were HIV negative, and three patients HIV status were unknown. 36(73%) patients had no co-morbidities. 6(12%) patients had Diabetes Mellitus. Twelve (24%) cases had pulmonary TB involvement.Forty three out of 49(86%) had back or neck pain as their primary symptom at presentation. Thirty-one (63%) patients had constitutional symptoms. Four (8%) patients had spinal cord compression, two patients had paradoxical drug reactions and 17(34%) patients needed surgery/or drainage of paravertebral or psoas abscesses.All patients had MRI whole spine performed at diagnosis. 19/49 (38%) had multiple spinal level involvement. The most common level affected was the thoracic spine 35/49 (71%), followed by lumbar 25/49 (51%), cervical 10/49 (20%) and sacral spine 7/49(14%). 19/49 (38%) had paravertebral abscesses, 5/49 (10%) had psoas abscesses.All patients had a repeat MRI spine at around 6 months (range 3 to 8 months) and 12 months (range 10 to 14 months). Twenty-eight out of 49 (56%) patients had persistent paravertebral and psoas abscesses at 6 months. All patients received 12 months of TB treatment and the majority of patients, 46/49 (92%) had complete resolution of paravertebral or psoas abscesses after 12 months of treatment.Abstract P136 Figure 1ConclusionPatients with spinal TB should have 12 months of treatment compared to 6 months as recommended by NICE guidelines.

IntroductionRecent trial data have demonstrated that both Drug-sensitive (DS) and Drug-resistant (DR) tuberculosis (TB) can be treated effectively with short-course quinolone-containing combination therapy. Current WHO guidance includes using a four-month treatment with quinolones throughout for DS TB. The adverse event profile of quinolones is well-recognised, however less is known about the prevalence of quinolone resistance, in particular when used to treat otherwise DS TB. The routine use of M tuberculosis complex Whole Genome Sequencing (WGS) in England, provides an opportunity to explore this in both DS and DR TB.MethodsWGS isolate data for M tuberculosis complex between 2017 and 2023 were obtained from NMRS-North and Central (N & C) and NMRS-South (S) UKHSA laboratories. After data cleaning, analyses were performed separately and then combined (data presented here). These focused on the relationship between quinolone, isoniazid and rifampicin resistance.ResultsIsoniazid resistance was present in 7.9% of 15350 isolates. Rifampicin or Rifampicin + Isoniazid resistance was seen in 2.1%, and WGS quinolone resistance was detected in 1.5% of. However, a greater proportion of WGS results for quinolones showed a mutation of uncertain significance (‘unknown’) than for other drugs (table 1, with all NMRS data combined and Isoniazid WGS profile included for comparison, percentages are for each row). Reassuringly, on phenotypic testing only 4/1021 (0.4%) NMRS-S isolates reported as ‘unknown’ on WGS were quinolone resistant.Column 3 of table 1 shows that quinolone resistance was present in 1% and 7.5% of Isoniazid DS and DR isolates.Abstract P141 Table 1 Isoniazid Quinolone Sensitive Quinolone Resistant Quinolone Fail Quinolone Unknown TOTAL Sensitive 11356 (85.2%) 119 (0.9%) 216 (1.6%) 1632 (12.2%) 13323 (100%) Resistant 974 (80.2%) 91 (7.5%) 23 (1.9%) 126 (10.4%) 1214 (100%) Fail 142 (64.5%) 3 (1.4%) 58 (26.4%) 17 (7.7%) 220 (100%) Unknown 468 (78.8%) 13 (2.1%) 6 (1%) 107 (18%) 594 (100%) DiscussionQuinolone-resistant M tuberculosis is present at a low though detectable frequency in all TB populations. The large number of WGS unknown results for quinolones that require further phenotypic testing is an additional time and financial cost that needs to be considered when implementing programmatic short-course treatment to adults with drug sensitive TB in the UK.

Introduction and ObjectiveThe COVID 19 pandemic caused significant disturbances in TB diagnostic and treatment services under national tuberculosis elimination programme (NTEP) in India. As India accounts for 28% of global burden, ending Tuberculosis globally is critically dependent on ending it in India. More than quarter of the world’s 10 million estimated cases and 449,700 of the world’s estimated 1.3 million TB related deaths occur in India. Between 2020 and 2025, 6 million TB cases and 1.4 million TB-related deaths are expected to occur in India. Due to lockdowns or movement restrictions, fear of contracting COVID-19 infection in hospital settings and diversion of TB services, patients with TB symptoms are having difficulty accessing healthcare facilities during this epidemic. The Objective of this study is to identify the real-world practical difficulties faced by TB patients during the COVID 19 pandemic during the second wave from March 2021 to October 2021 in India. MethodsFigure 1.ResultsOut of 100 patients diagnosed with drug sensitive Tuberculosis, 42% were COVID-19 suspects.38% had symptoms for less than one month which helped in early diagnosis of Tuberculsosis.6% patients had symptoms for more than 6 months.27% patients faced problems getting diagnosed, out of which 14 patients (51.8%) had travel difficulty, 7 patients (25.9%) had financial problems and 6 patients (22.2%) had lack of health care access. The time taken for diagnosis and starting medication under National TB elimination program (NTEP) was 1–3 days in 47% patients, 4–7 days in 32% patients and 8 or more days in 21% patients. 31% of patients had side effects due to anti-tuberculosis treatment, amongst them 23 (74.1%) patients complained vomiting, 5 (16.1%) patients had itching, 3 patients (9.6%) had joint pains. 84% patients received regular supply of anti-tuberculosis medication and 16% faced problems in access.79% patients had access to high protein diet whereas 21% patients had no access. Abstract P143 Figure 1ConclusionThis study highlights the consequences and impact of the COVID-19 pandemic on the Tuberculosis healthcare services. It highlights the problems faced during the COVID-19 lockdown by Tuberculosis patients.

IntroductionRifampicin plays a key role in tuberculosis (TB) treatment, due to its bactericidal and sterilizing capacity. The dose currently recommended by WHO is 10 mg/kg but achieving this dose can be limited by fixed dose preparations and risks under dosing. Low plasma levels of rifampicin may potentially contribute to poor treatment response. Therapeutic drug monitoring (TDM) has been proposed as a method to support early optimisation of dosing particularly in high-risk groups e.g. CNS TB or slow treatment response. The main aim of this study was to investigate whether standard dosing of rifampicin is likely to lead to therapeutic drug levels and to describe the impact of dose escalation.MethodWe performed a retrospective review of all adult patients who had rifampicin levels (2 hours post dose or 6 hours where delayed absorption was suspected) taken between August 2018 and August 2021, identifying 58 patients. Patient demographics, clinical characteristics including weight at time of level, dose adjustments and rationale for taking level were collected from patient records. Levels below 8 mg/L were considered subtherapeutic.ResultsThe most frequent reason for conducting a rifampicin level was due to slow/poor response (34%) followed by high burden of disease (24%). Just under half of patients (48%) were being treated for pulmonary TB, followed by extrapulmonary TB (24%). Of the 58 patients who had levels measured, 40 (69%) patients had levels that were considered to be subtherapeutic. Figure 1 displays if therapeutic levels were achieved with standard rifampicin dosing (10 mg/kg). In the subtherapeutic group, 31 (78%) had their dosage increased of which 19 (61%) patients had a level taken post dose escalation. Eleven 11 (58%) patients from this group had levels within therapeutic range.Abstract P142 Figure 1Percentage of patients attaining therapeutic levels based on initial rifampicin dose (mg/kg)ConclusionBased on our finding’s a significant proportion of patients show subtherapeutic levels on standard dosing particularly in those with apparent severe or poorly responding disease. TDM is a useful tool to individualise rifampicin dosing and early optimisation increases the likelihood of attaining a therapeutic level. This may be particularly beneficial in patients who may be at risk of low plasma levels e.g. malnourished/disseminated/CNS TB.

Hormonal contraception is known to precipitate or perpetuate depression in some patients. The link between oral contraceptive pills and depression relates to the amount and type of progestogen contained in these pills. Many of the older oral contraceptive pills, which contain ethinylestradiol, are linked to severe mood problems. Newer oral contraceptive pills containing physiological forms of oestrogen may be better tolerated with a purported weaker link to mood problems. Clinicians should consider the temporal relationship between the use of hormonal contraception and development of new or worsened depression or mood changes.