Treatment with opiates and benzodiazepine increases mortality risk

14th December 2020

While rates of opiate prescribing in the US has levelled off, it appears that co-prescription with benzodiazepines is on the increase.

Although long-term benzodiazepine use is associated with an increased mortality risk, whether this risk is greater when combined with opiates remains unclear.

Both insomnia and anxiety are often managed through prescription of benzodiazepines. However, these drugs are associated with several adverse effects including an increased the risk of falls, cognitive impairment and a life-threatening risk when being withdrawn. Though some evidence suggests that benzodiazepine use alone increases mortality risk and that co-prescription with opiates is also potentially lethal, the data is inconsistent.

As a result, a team from the Health and Behaviour Research Centre, Department of Psychiatry, Washington University, US, turned to data available in the National Health and Nutrition Surveys (NHANES) to examine the association between benzodiazepine use, either alone or in combination with opiates and mortality. Using the NHANES data set collected between 1999 and 2104, researchers obtained a nationally representative sample of approximately 5000 participants for whom mortality data was available through to 2015. Eligible patients were 20 years and older and data on prescription use was also obtained from NHANES as were potential covariates for the analysis including other medical and psychiatric conditions which may have been associated with an increased risk of mortality. In addition, a cohort of patients prescribed selective serotonin re-uptake inhibitors (SSRIs) were included as a control group.

Findings
The sample contained 5212 participants (38.2% male) with a mean age of 54.8 years of whom, 468 (9%) were co-prescribed benzodiazepines and opiates, 1256 (24.1%) benzodiazepines alone and 1955 (37.5%) opiates alone. In addition, 1533 were prescribed neither benzodiazepines or opiates but SSRIs. After a median follow-up of 6.7 years, there were 101 deaths in those receiving co-treatment, 236 among benzodiazepine only and 227 deaths among SSRI users. After adjustment for covariates, there was a significant increased risk of death for co-treated individuals (hazard ratio, HR = 2.04, 95% CI 1.65 – 2.52) and although smaller, also a significant risk for those using benzodiazepines alone (HR = 1.60, 95% CI 1.33 – 1.92). Interestingly, the researchers observed a significant increased death risk for co-treatment only among those aged under 65 years of age (HR = 3.27, 95% CI 2.40 – 4.47) and among those taking benzodiazepines alone (HR = 1.81).

They concluded that co-treatment is associated with a twofold increased risk of mortality and called for a concerted effort among physicians and policy-makers to reduce prescribing of the medications and to identify those at an increased mortality risk.

Citation
Xu KY et al. Association between benzodiazepine use with or without opioid use and all-cause mortality in the United States, 1999-2015. JAMA Netw Open doi:10.1001/jamanetworkopen.2020.28557

AZD1222 vaccine has acceptable efficacy and safety profile

The highly anticipated vaccine (AZD1222) produced by a collaboration between Oxford University and AstraZeneca has become the first to have its interim study data published and appears to be as effective as reported in various media outlets.

AZD1222 is a viral vector based on an adenovirus derived from chimpanzees and subsequently modified so that it would induce a strong immune response against COVID-19. An interim analysis of the data has now been published by the Oxford Vaccine Group, based on four ongoing, and randomised controlled trials done across three countries. The trials include a Phase I/II (COV001) and Phase II/III (COV002), both from the UK, together with a Phase III trial (COV003) conducted in Brazil and a Phase I/II study (COV005) in South Africa, although three of these trials are single blind and only the South African study is double-blind. The interim data comes from all patients in the trials who received two doses of AZD1222. In COV001 patients are aged between 18 and 55 and randomised 1:1 to the test vaccine or a control vaccine (meningococcal group A, C W and Y). In COV002, participants (also aged 18 to 55) have received two doses of the vaccine, with the first being a lower dose which was boosted by the second, standard dose. However, subsequent enrolment to this arm occurred with patients aged 56 – 69 years of age and at a later stage patients aged 70 years and over and this later cohort received two standard doses of the vaccine.

The authors also reported that while COV002 was originally designed as a single dose study, this was amended to include a second booster dose. COV003 recruited individuals at high risk e.g., healthcare workers, 18 years and older and received a different dose again of the test vaccine administered up to 12 weeks apart. Finally, COV005 included healthy adults (18 – 65 years of age) who received the same two doses received in COV003. For the interim analysis, all studies were required to have at least 5 cases eligible for inclusion but neither COV001 or COV005 met these criteria and hence the interim analysis was based on only two studies. Vaccine efficacy was calculated 1 – adjusted relative risk and the primary outcome was set as virologically confirmed COVID-19 (via a swab test) combined with at least one qualifying symptom (e.g. fever, cough, shortness of breath or anosmia or ageusia.

Findings

A total of 23,848 patients were recruited across the 4 trials and 11,636 were included in the interim analysis. The majority of patients (86%) were aged between 18 and 55 and 12.2% were older than 56 years (those recruited later in the trial). In the trial with an initial lower dose, there were 33 cases of COVID-19, 3 (0.2%) in the test vaccine group and 30 (2.2%) in the control arm, which gave a vaccine efficacy of 90%. among those who received two standard doses, vaccine efficacy was 62.1%.

The authors were unable to explain why a lower followed by a standard dose resulted in higher vaccine efficacy and this is the subject of further work.

Citation

Voysey M et al. Safety and efficacy of the ChAdOx1 nCOV-19 vaccine (AZD1222) against SARS-CoV-2; an interim analysis of four randomised controlled trials in Brazil, South Africa and the UK. Lancet 2020 doi.org/10.1016/ S0140-6736(20)32661-1

Approximately a fifth of people across the world would refuse a COVID-19 vaccine

As a COVID-19 vaccine is rolled out across the UK and the US, a meta-analysis of international studies suggests a sizeable proportion of people are unwilling to be vaccinated which has important implications for the effectiveness of a vaccination programme.

It is fairly well established that a route out of the restrictions imposed on individuals because of COVID-19 would be through widespread vaccination once a suitable product becomes available. Early in the pandemic, several studies sought to ascertain people’s perceptions on whether they would be willing to receive a vaccination against the virus. Such information is of vital importance because it is believed that a large proportion of a country’s population needs to be vaccinated for the program to be fully effective. The earliest data was positive, with a large majority of those surveyed, stating that they were likely to want to be vaccinated. However, given the pace with which vaccines have entered developed, there has been a good deal of misinformation circulated on social media, leading to potential mistrust in the safety of vaccines which has been produced and tested in such a short space of time than normal.

Given this potential uncertainty, a team led by researchers from the Department of Psychology, University of Liverpool, set out to systematically review all the available studies undertaken to examine the percentage of the population intending to be either vaccinated or intending to refuse it once available. Included studies were required to have used a nationally representative sample (e.g. based on age, gender, education level etc) and having a sample size greater than 1000 participants from the same country though any studies from non-general public samples e.g., healthcare professionals, students, were excluded. In addition, all eligible studies had to include a question that measured willingness to use a vaccine for COVID-19 once available and these studies had to report the outcome for each option, e.g., yes vs no, or willing vs unsure vs unwilling.

Findings
A total of 20 articles reporting on 28 samples were included which were undertaken in 13 different countries. The size of each sample varied from 1,000 to 7,547 with a median of 1,198. In addition, samples were collected in the early phase of the pandemic (March – May 2020) or later (June 2020 and onwards). Interestingly, the proportion of respondents willing to be vaccinated decreased over time (79% early phase studies vs 60% later studies), whereas the proportion not willing to be vaccinated increased (12% early studies vs 20% later studies). In contrast, the proportion of individuals who stated that they were unsure did not change over time. There was also a persistent trend relating to vaccination intentions: being female, younger, of lower income/education level or belonging to an ethnic minority were all associated with a reduced likelihood of wanting to be vaccinated.

The authors called for an urgent need to address vaccination hesitancy to promote widespread uptake of the vaccine.

Citation
Robinson E et al. International estimates of intended uptake and refusal of COVID-19 vaccines: a rapid systematic review and meta-analysis of large nationally representative samples. MedRxiv 2020 doi.org/10.1101/2020.12.01.20241729

Topical steroids normalise inflammatory markers in infants with atopic eczema

In atopic eczema, immune dysfunction occurs both in the skin and systemically, although whether treatment with topical steroids rectifies this dysfunction at both sites is unknown.

Atopic eczema is a chronic, inflammatory skin condition that affects up to 20% of children. The condition usually begins within the first year of life and treatment involves the regular use of emollients and disease flares are managed with topical steroids to dampen down the localised inflammatory response. Previous work has revealed how immune dysfunction in the skin is characterised by the production of pro-inflammatory cytokines but that there are also changes in the level of systemic markers during a disease flare.

In an attempt to gain a better understanding of how these markers change during therapy with topical steroids, researchers from the Department of Paediatric Dermatology, Children’s Health at Crumlin, Dublin, Ireland, examined the changes in cytokine markers in an observational study of children with atopic eczema experiencing a disease flare treated with the topical steroid, hydrocortisone 1%, in line with normal clinical practice guidance. All children were less than a year old with moderate to severe eczema based on a SCORAD (disease severity) score of 25 or greater. Team also included a control group of healthy children to compare any changes in cytokine levels. Samples of the skin and plasma were taken and both analysed for a large number of different biomarkers including inflammatory cytokines such interleukin-18 (IL-18), IL-5, IL-13, CXCL8, il-1 alpha and CCL17 and CCL22.

Findings
A total of 47 children with atopic eczema and 20 controls were included in the analysis and changes in biomarkers were compared between baseline and after 6 weeks of topical steroid therapy. The most pronounced differences in cytokine levels in the skin between eczema and healthy controls were for those associated with innate immune activation, e.g, IL-18, CXCL8, IL-1alpha, CCL17 and CCL22 although levels in eczema patients did not reduce to the levels observed in healthy controls. In contrast, the largest changes in the blood were with CCL17, IL-13, CCL22, IL-5 and CCL26. Interestingly, these latter changes, which represent key cytokines in the pathophysiology of atopic eczema, only occurred in the blood and not in the skin.

Commenting on their results, the authors suggested that their data indicates that topical steroids had both a local and systemic effect to minimise the inflammation response induced by an eczema flare.

Citation
McAleer MA et al. Topical corticosteroids normalise both skin and systemic inflammatory markers in infant atopic dermatitis. Br J Dermatol 2020 doi: 10.1111/bjd.19703

Real-world data show that biologics for psoriasis are less effective in practice

7th December 2020

Randomised controlled trials (RCTs) define the expected efficacy of a drug. However, real-world studies serve to demonstrate the effectiveness in clinical practice, which might be very different.

Real-world data can be derived from patient registries and are of value because they illustrate whether the expected efficacy determined in a randomised trial holds up in practice. Using information held in the British Association of Dermatologists Biologics and Immunomodulators Register (BADBIR), a team from the Centre for Dermatology, Salford, UK, examined the comparative effectiveness of two biologics, secukinumab and ustekinumab used in patients with moderate to severe psoriasis, defined in terms of a psoriasis and area severity index (PASI) of 12 or more prior to starting with either biologic. The PASI serves as a measure of disease severity with scores above 10, indicating at least moderate disease severity. The researchers restricted the drug initiation dates to on or after September 2013 (when both drugs became available) and before September 2018, to allow for patients to have completed at least 12 months of therapy. The primary outcome for the study was the difference in the proportion of patients who achieved a PASI of 2 or lower (i.e. virtually disease free) after 12 months of therapy. The authors used the results of a recent RCT in which the two biologics were studied head-to-head, as a baseline to compare the effectiveness data derived from the BADBIR registry.

Findings
A total of 1231 patients were included, 917 receiving ustekinumab (mean age 45 years, 40% female) and the remainder secukinumab (mean age 46 with 38% female). A PASI at 12 months post-treatment was available for only 42% and 45.5% of those given secukinumab and ustekinumab respectively. Secukinumab was superior to ustekinumab in achieving a PASI of 2 (relative risk = 1.28, 95% CI 1.06 – 1.55). However, a further and important finding from the BADBIR registry data was that the estimate of efficacy from the RCT appeared much lower in practice. For example, secukinumab and ustekinumab were 17.5% and 15.1% less effective than the RCT data would suggest.

Commenting on these findings, the authors suggested that given this discrepancy, clinicians should aim to inform patients prescribed either drug that the true effectiveness was likely to be lower than expected based on the results shown in the clinical trial.

Citation
You ZZN et al. Randomised trial replication using observational data for comparative effectiveness of secukinumab and ustekinumab in psoriasis. A study from the British Association of Dermatologists Biologics and Immunomodulators register. JAMA Dermatol 2020

Differences in immune and clotting functioning may explain less severe COVID-19 in children

The lower incidence of severe COVID-19 in children has prompted much speculation. In reviewing various hypotheses, researchers believe that changes in endothelial cells and clotting functions is a likely reason for the disparity.

A consistent finding of infection with COVID-19 is that the virus is less severe in children, most of whom are asymptomatic, with a mild fever, cough and changes in taste or smell. Whether or not children are less likely to become infected is an ongoing debate but overall, only between 1 and 2% of cases involve children, suggesting that as a group, children are less likely to become infected, even when exposed to similar viral loads.

In a review of all the currently available evidence, a team from the Faculty of Science and Medicine, University of Fribourg, Switzerland, have examined the various potential physiological differences between adults and children and how these might account for the variation in rates of infection.

While there are still no definite answers, the authors suggest that infection of endothelial cells, leading to vasculitis and the formation of microthrombi, is more problematic in adults, especially in the presence of comorbidities such as hypertension and diabetes, both of which affect endothelial cell functioning. While children are less likely to have comorbidities and are therefore at lower risk, immunosuppressed children or those with cancer, do not appear to be at any higher risk from COVID-19. Another possible protector is that children have a stronger innate immune and adaptive immune system with higher number of natural killer cells and both B and T cells, whereas ageing is associated with immunosenescence, i.e., a gradual decline of both immune systems. Other proposed differences which might be protective in children are higher levels of melatonin and the microbiota in the nasopharynx in children which is more heavily colonised in children compared with adults.

The authors conclude that with the exception of the endothelial/clotting function disparities, none of the currently proposed hypotheses are able to account for the differences in susceptibility to infection between adults and children.

Citation
Zimmerman P, Curtis N. Why is COVID-19 less severe in children? A review of the proposed mechanisms underlying the age-related difference in severity of SARS-CoV-2 infections. Arch Dis Child 2020 doi:10.1136/archdischild-2020-320338

Nutritional supplements potentially beneficial in COVID-19

During the COVID-19 pandemic, sales of nutritional supplements known to have an important role in immune function such as vitamins C, D and the mineral, zinc, increased substantially. However, to date, no trials have shown evidence for a beneficial role of these supplements in COVID-19.

Given this lack of evidence on the possible value of nutritional supplements in either preventing or reducing the severity of infection with COVID-19, researchers from Kings’ College, London, wondered if they could examine the association between supplement use and self-reported infection with the virus, from data collected as part of the COVID Symptom study app. The app allows users to self-report a vast amount of data including demographics, core risk factors, COVID-19 symptoms and test results. In fact, data collected from the app has already led to the acceptance of anosmia and skin rashes as important symptoms of COVID-19 infection. Using a yes/no question, the app enquires as to whether an individual uses any nutritional supplements on a regular basis, defined as more than three times a week. The app has a list of various supplements including probiotics, garlic, omega-3 fish oils, multivitamins as well as vitamins C, D separately and the mineral, zinc. For the present analysis, the team included a subset of participants who reported being tested for COVID-19 using either the PCR test or via serology and who reported that they used nutritional supplements. The primary outcome was the report of a positive test for COVID-19.

Findings
A total of 175,652 individuals self-reported using nutritional supplements regularly since the beginning of the pandemic. Analysis of their data showed a significant 13% reduced risk of testing positive for COVID-19 and regular use of multivitamins (odds ratio, OR = 0.87), supplementing with vitamin D (OR = 0.91), probiotics (OR = 0.86) and omega fish oils (OR = 0.88). In contrast, there was no statistically significant association for either zinc, vitamin C or garlic. Interestingly, these protective effects were only significant for females and this was seen across all age groups and body mass indices. No protective effect was seen for males although for males aged > 60 years, there was a positive association between use of zinc and vitamin C and testing positive for COVID-19.

Although the associations were not huge, the researchers called for trials to test the protective effects of these nutrients before making any specific supplement recommendations.

Citation
Louca P et al. Dietary supplements during the COVID-19 pandemic: insights from 1.4M users of the COVID Symptom Study app – a longitudinal app-based community survey. MedRxiv preprint doi: https://doi.org/10.1101/2020.11.27.20239087

Medicinal cannabis does not impair driving performance

4th December 2020

Epidemiological studies have shown that cannabis is associated with an increased risk of car crashes

Whether or not, cannabidiol, the derivative widely used as medicinal cannabis, also impairs driving performance is largely unknown.

Cannabis is available in three different varieties (or technically, chemovars) which are cannabidiol (CBD) dominant, tetrahydrocannabinol (THC) dominant and a CBD-THC equivalent mixture. Individuals inhale the THC dominant form because of its intoxicating effects. Whether the CBD dominant form, used medicinally, has sedating properties is currently equivocal.

As a result, researchers from the Lambert Initiative for Cannabinoid Research, University of Sidney, Australia, conducted a randomised trial to examine the extent and duration to which each of the different cannabis chemovars impaired driving performance. They recruited 26 individuals with a history of occasional cannabis use and employed a cross-over design study that involved four on-road driving periods, one week apart.

For the study, the participants inhaled one of vapourised cannabis containing 13.75mg CBD, THC, CBD-THC equal mixture or placebo. The driving tests twice; 40 to 100 minutes and 240 to 300 minutes post-vaporisation. The primary outcome was the mean standard deviation of lateral position (SDLP), which measures the extent of lane weaving, swerving and overcorrecting.

Findings
The mean age of the 26 individuals was 23.2 years (including 16 women), all of whom completed the 8 driving tests. At 40 to 100 minutes and following inhalation of cannabis, the mean SDLP after CBD was 84.07cm, 86.94cm after THC, 84.07cm after the CBD-THC mixture and 84.41cm after placebo and these values remained similar between 240 and 300 minutes.

Pairwise comparisons revealed that both SDLP values for THC and the THC-CBD were higher than placebo at 40 to 100 minutes, but not between 240 and 300 minutes. In other words, both THC and the combination but not, CBD, impaired driving performance in the short-term, but that this effect was abolished in the longer term.

Although a limitation in the study, recognised by the authors, was that the dosages used might not reflect common usage, their data implied that CBD did not affect driving ability.

Citation
Arkell TR et al. Effect of cannabidiol and delta-nine tetrahydrocannabiniol on driving performance. A randomised clinical trial. JAMA Net Open 2020 doi:10.1001/jama.2020.21218.

Low BMI increases risk of death with COVID-19

2nd December 2020

Obesity is associated with a worse prognosis among individuals with COVID-19. However, until now, no studies have examined the relationship between different body mass indices and clinical outcomes.

Preliminary data has suggested that being overweight can contribute towards poorer health outcomes for those with COVID-19 and such patients are more likely to require mechanical ventilation. In contrast, far less attention has been paid to the possible relationship between low body mass index (BMI) and clinical outcomes. As a result, a team from the Department of Medicine, Donald and Barbara Zucker School of Medicine, New York, USA, set out to retrospectively analyse the overall association between different levels of BMI and outcomes for patients with COVID-19. The team included 10,861 patients admitted with COVID-19 into their hospital during March to the end of April 2020. The BMI was documented on admission to hospital as either self-reported or obtained from nursing staff. It was classified as follows with the BMI range in brackets: underweight (under 18.5kg/m2), normal (18.5 to 24.9), overweight (25 to 29.9) and obese class I (30 – 34), II (35 – 39) and III (40 or more). Data on patient outcomes were collected from electronic health records and the primary outcome used was the need for mechanical ventilation and death. All patients were followed until discharge, death or until early May 2020.

Findings
For the 10,861 patients with a median age of 65 years (59.6% male), 2.2% were classed as underweight, 23.1% normal, 37% overweight and 37.7% obese. In total, 2,220 (20.4%) patients required mechanical ventilation and 2,596 (23.9%) died. There were significant differences in both the need for mechanical ventilation and death among the different BMI categories. For instance, 12.4% of those classed as underweight and 73.8% of those classed as obese (all classes) required mechanical ventilation. In addition, 39.2% of those who were underweight died as did 62.9% of those in the obese category. Regression analysis revealed that patients who were underweight an increased risk of death (hazard ratio, HR = 1.46, 95% CI 1.17 – 1.81) which was actually higher than the risk for those with class III obesity (HR – 1.23, 95% CI 1.03 – 1.48). The reason for the increased risk of death among underweight patients was unclear but the authors speculated that this may have been related to increased frailty.

They concluded that the findings suggest that worse outcomes after COVID-19 are not restricted to obese patients and those who are underweight are also at a higher risk.

Citation
Kim T et al. Body Mass Index as a Risk Factor for Clinical Outcomes in Patients Hospitalised with COVID- 19 in New York. Obesity 2020 doi.org/10.1002/oby.23076

Latest REACT-1 study data suggest COVID-19 rates are falling

1st December 2020

Data from the latest cross-sectional survey of COVID-19 from a random sample of people in England suggest that the prevalence of infection is decreasing.

The REal-time Assessment of Community Transmission-1 (REACT-1), which was established by a team at Imperial College, London, has been collecting monthly and random data since May 2020. The study is designed to examine time trends in the prevalence of positive, self-administered nose and throat swab tests in England. The collected data can then be used to estimate the prevalence of COVID-19 by age and in different regions of the England.

Findings
The latest data show that there were 821 positive cases from 105,123 swab tests (0.78%), giving a weighted prevalence of 0.96% This is approximately 30% lower than the figure observed during the last period (1.32%) collected between 26 October and 2 November 2020. Using these results, the team have estimated that the R value is 0.88. The information on positivity for the different regions of England show that there has been a reduction of more than 50% in the North West and North East (where levels were previously much higher) though little change in the East and West Midlands and London. Based on their findings, the authors estimate a national prevalence of 0.96%, equivalent to approximately 720,000 infections in England on any one day.

However, the study is not all good news. There appears to be an increased prevalence in those aged 5 to 17 years, i.e., school children. Additionally, the data show that there is no evidence of decline among people of Black and Asian ethnicity, with the latter group having a much higher odds of testing positive (odds ratio = 1.72) compared with White individuals. There were also increased odds of testing positive among those from the most socially deprived areas and among health and care workers.

The authors conclude by noting that while the prevalence of infection appears to be reducing, it is still high at roughly 1% of the population and that measures will be needed to suppress levels further such as widespread vaccination.

Citation
Imperial College. REACT-1 round 7 interim report: fall in prevalence of swab-positivity in England during national lockdown.