Clinical & Policy Updates:
SMMGP Clinical Update #40 - May 2014
Compiled by Dr Euan Lawson
|Download the PDF version of this Update here! (PDF*, 82K)|
Substance, structure and stigma: parents in the UK accounting for opioid substitution therapy during the antenatal and postnatal periods
Chandler A, Whittaker A, Cunningham-Burley S, et al. Int J Drug Policy 2013;24:e35-42
This paper reports on a Scottish qualitative study that interviewed 19 opioid-dependent service users (14 women and 5 men) who were all expecting, or had recently had, a baby. The aim was to explore the impact of opiate substitution therapy (OST) on parents and their experience of parenting support.
The authors described how the participants were orientated toward demonstrating that they were doing "the best thing" for their baby. Opiate substitution therapy was seen by some as a route into normal family life - but it was also perceived by others as being a barrier. Important themes included: the physiological effects of OST; the structural challenges of treatment regimes and medication collection; and the impact of society's negative views.
Commentary: This can be an area of practice mired in tragedy and sadness. This paper tries to cut through some of the raw emotion, the blind prejudice, and ambiguity that often clouds discussions around parenting for people on OST. It's all horribly complex and trying to pick out a single key point from a paper like this is nearly impossible. Attitudes to OST vary hugely and that is, perhaps, the main message from this study - it clearly points out how treatment-by-numbers is a crackpot fantasy of policy makers. All we can do is consult on an individual basis and take account of the person's own beliefs and desires. One quote highlights the anxieties people have about negative associations with OST - but also the unclouded clarity of thinking children can display:
"I've ta'en her [older child] to the chemist with me a couple o' times... and I dinnae like daein it cos I'm still on supervised. But she's just, "Ah, mum, get your medicine, are you all better now?" But it's no' something I like daein in front of her..."
I often think a decent qualitative paper like this is worth a dozen meta-analyses when it comes to managing day-to-day consultations. It would be hard to read this paper and not call it to mind the next time you see a pregnant woman or a new mum and their partner.
Treatment retention, drug use and social functioning outcomes in those receiving 3 months versus 1 month of supervised opioid maintenance treatment. Results from the Super C randomized controlled trial
Holland R, Maskrey V, Swift L, et al. Addiction 2014;109:596-607
The aim of this study was to find out whether supervision had any effect on retention. It was a randomised control trial based in three areas of England - specifically Hertfordshire, Norwich and London. The four community drug services involved were running services over 12 clinic sites. The intervention group was expected to receive supervised treatment for three months - either six or seven days per week depending on pharmacy opening on the Sunday. The unsupervised group was, as one would expect, supervised during the 7-28 days it took for dose titration and stabilisation. Although supervision was then stopped they still had daily pick-ups at the pharmacy. The primary outcome was retention in treatment at 3 months. Secondary outcomes included: retention at 6 months and time to "drop out" of treatment; reduction in use of illicit opioids; quality of life; and criminal behaviour.
They recruited a total of 293 who were deemed appropriate to enter the study - people could be excluded if a physician felt they needed supervision. A total of 201 people (32%) of 627 assessed were excluded on this basis - further details on this group were not available. After randomisation there were 145 in the supervised group and 148 in the unsupervised. At 12 weeks, 69% of the supervised group were in treatment compared with 74% of the unsupervised (non-significant difference). Retention was similar at 6 months in the two groups. There was no significant difference in the survival analysis - for retention that is (not actual deaths). The use of illicit opioids dropped in both groups. Both groups showed improvements in quality of life measures but there was no difference between the groups. The results showed that 21% of supervised patients committed crime versus 9% unsupervised (OR 3.37, 95% CI 1.28-8.86).
Commentary: There is indirect evidence that supervision has led to reduced deaths from methadone toxicity. The Orange Book guidance for a 12 week period of supervision has always seemed relatively arbitrary - but much has been made, quite rightly, of the observational evidence that overdose deaths have dropped impressively in the period when this became usual UK practice. The Super C randomised controlled trial is an important UK-based study that tries to tease out some of the unknowns.
The authors did an intention-to-treat analysis (ITT) but they also looked at the results without this - the so-called "per protocol survival analysis" suggests that the supervised patients were less well retained but the odds ratio for this was 0.71 (95% CI 0.51-1.00). Given this isn't on an intention-to-treat basis and given the confidence interval actually touches 1.00 this should be treated with deep caution.
What this paper really shows is that there was really very little difference and that's important in itself. It's difficult to know how much stock to put in the finding that the supervised patients committed more crime - it could simply be a chance finding. There are also some unfortunate confounders that have crept in - there is little the authors could have done about this - but randomisation resulted in several significant differences between the two groups. The supervised group had more women, more unemployed, fewer criminal convictions, fewer previous methadone scripts, fewer had a current physical diagnosis, and fewer of them used alcohol.
One could draw the conclusion that one month is as good as three when it comes to supervision. However, another key aspect of this study is the people who were excluded at the eligibility stage of the study. A third of the potential participants were excluded from the study because a doctor decided they couldn't be unsupervised. That may have been felt to be an important safety mechanism in this study but it's a very value based exclusion criterion and one wonders on the practical ramifications of this - it certainly should affect how clinicians interpret the results. This study is described as a "pragmatic" RCT - given the nature of the supervision it would be challenging for it to be otherwise. It's a good study of an intervention that deserves some scrutiny.
Factors associated with recent symptoms of an injection site infection or injury among people who inject drugs in three English cities
Hope VD, Hickman M, Parry JV, et al. Int J Drug Policy 2014;25:303-7
This paper reports on a study that recruited patients using "respondent driven sampling" (RDS) in Bristol, Leeds and Birmingham. RDS involved recruiting via street outreach and key informant referrals. The participants took part in a computer-assisted interview and provided a dried-blood sample. The participants (all injectors within the previous 4 weeks) were asked if they had experienced any symptoms at an injection site. The samples were tested for hepatitis B core antigen and HCV antibodies.
In total they recruited 855 people. The two most common injection sites were the arms (41%) and the groin (37%). In the preceding 28 days, 21% reported redness, swelling or tenderness at an injection site, 6.1% reported an abscess and 5.2% reported a sore/open wound. In the multivariate analysis, those who reported redness, swelling or tenderness were more likely to have been arrested in the past year, injected more frequently, and used multiple injection sites. It was less frequent in those who used their groin as their main injection site and who reported always cleaning their spoons/mixing containers (71%).
An abscess was more common in those who reported an overdose in the past year, who injected daily, were using multiple injection sites, and who injected into their legs. An abscess was less common in those who always swabbed their injection sites (52%).
An open sore/wound was more commonly reported by women, those who injected daily, were using multiple injection sites, and those who reported their main source of income as being illicit.
Commentary: The high prevalence of injection site infections won't come as a surprise to a primary care readership. The absence of research around this topic is more surprising. Some of the factors here are relatively obvious: people who inject drugs more frequently are more likely to run the risk of infections. In addition, it's easy to envisage how the injecting will get more risky as good veins dwindle. There is some genuine pointers towards clinical advice we can give to people: women, and those injecting into their legs seem to have a particular vulnerability. Advice around swabbing injection sites and using clean spoons/mixing containers could be warranted given the reduced problems for these people - though it could simply represent individuals who are somewhat more fastidious when it comes to injecting.
Case-finding for hepatitis C in primary care: a mixed-methods service evaluation
Datta S, Horwood J, Hickman M, et al. Br J Gen Pract 2014;64:e67-74
This UK study aimed to identify patients at risk of developing hepatitis C using routine GP data and also to work out how many had not been tested. In addition, it explored the views of GPs in regards to this. It was based in six NHS practices in Bristol. The quantitative aspect of the study involved searching on EMIS LV and EMIS Web using a defined set of Read Codes. These included: injecting or intranasal drug use; born or brought up in a country with high HCV prevalence; blood transfusion before 1991; blood products prior to 1986; transplant pre 1992; infection with HIV; infection with HBV; and born to a mother with HCV. Test results were obtained from the local HPA laboratory. The qualitative element to the study involved inviting GPs for interview. One researcher conducted 17 face-to-face interviews with 17 GPs.
Quantitative analysis: The results showed that out of a total of 73,814 patients at these practices there were 3765 (5%) who were identified as being at high risk. Out of those 3754 patients a total of 308 patients had tested positive (8%), 406 had tested negative (11%) and 3051 (81%) had no result. Out of the 308 who were HCV positive a total of 76% were people who inject drugs.
Qualitative analysis: Three key themes emerged from the qualitative analysis of the study: the motivation for HCV testing; barriers to HCV testing; and ideas for improvements to practice.
Can hepatitis C virus infection be eradicated in people who inject drugs?
Grebely J, Dore GJ. Antiviral Res 2014;104:62-72
Good question. And in summary: yes! The authors make the case that we should certainly be striving for the elimination of HCV and go on to describe how we might go about doing it. It should be pointed out that elimination is not eradication. Eradication is the complete and permanent worldwide reduction to zero new cases - elimination is the reduction to zero (of incidence) in a defined geographical area. Harm reduction measures such as needle and syringe exchange will need to continue to make any of this feasible. Treatment, particularly in groups where there is a high risk of transmitting infection, will require targeted programmes. Global effects will come to nothing unless we expand treatment access.
Commentary: The BJGP paper is a study in general practice in the UK to work out how to find people with HCV. It can barely get more relevant. Let us pause and consider one of the most important facts from this study - out of all those people who had clear indicators of a need for HCV testing a whopping 81% had no apparent result. Sure, some of these may have been done and not Read Coded (though one practice did handsearch the notes to check this), but the unvarnished message is that testing for HCV is pitifully low. A treatable disorder is being missed.
And the second commentary by Grebely and Dore emphasises the long-term goal here. Richard Lehman, the BMJ's medical update blogger, wrote on the BMJ website on the 14th April 2014 in response to the flurry of papers published in the NEJM on the new HCV medications: "A deadly virus has been conquered. Hepatitis C genotype 1 can be cleared with a simple oral combination treatment, and compared to that, the rest of this week's medical news seems minor." It's not just the news of the week - it could be the news of the year or decade. There can be genuine hope that HCV can be eliminated in PWID - perhaps it is over-optimistic but it is feasible. Few prizes in medicine are as great as the complete elimination of a pathogen from a vulnerable group - the benefits in terms of mortality, morbidity and quality of life could make you giddy.
The BJGP paper is the much-needed slap in the face that brings us back to reality - we still have a very long way to go.
Post-mortem whole body computed tomography of opioid (heroin and methadone) fatalities: frequent findings and comparison to autopsy. Winklhofer S, Surer E, Ampanozi G, et al. Eur Radiol Published Online First: 6 March 2014.
This was a study that took 55 people who had died and in whom heroin and/or methadone overdose had been found to be responsible for the death. They performed whole-body postmortem computed tomography (PMCT) in those cases and also in 55 age- and sex-matched controls.
The most common findings in the study group were pulmonary oedema (95%), aspiration (66%), distended urinary bladder (42%), cerebral oedema (49%), pulmonary emphysema (38%), and fatty liver disease (36%). All of these were significantly more common than in the control group. It was noted that the triad of lung oedema, brain oedema, and distended urinary bladder was seen in 26% of cases and no controls
Commentary: This may seem like an odd paper to include in a clinical update but there are some lessons here that can be drawn out for clinicians. And, as painful as it is, there is no question that we have patients who die from overdoses and we may need to speak to pathologists or attend inquests. The triad of lung and brain oedema with distended bladders is described as a "highly specific constellation" for opioid intoxication. That won't help the living too much but will be important for pathologists and coroners - though it should be noted that, while specific, it wasn't hugely sensitive. The findings may have something for the living too. Pulmonary oedema is more common in those who survive overdoses as well. I've lost count of the number of times I've banged on about respiratory and liver disease in people who use drugs - the CT scans showed a remarkable 38% showed signs of emphysema and 36% had fatty liver disease. Need I say more?
Low incidence of adverse events following varenicline initiation among opioid dependent smokers with comorbid psychiatric illness
Nahvi S, Wu B, Richter KP, et al. Drug Alcohol Depend 2013;132:47-52
This study identified 575 smokers among 690 opioid-dependent patients (83.3%) and assessed 82 courses of varenicline treatment prescribed to 70 smokers. They assessed the adverse events and effectiveness of varenicline in this group. They noted that there was a very high prevalence of cardiovascular risk factors and psychiatric co-morbidity in those prescribed varenicline. Out of the 82 courses, nine of them (11%) were stopped as a result of adverse reactions. There were no reports of suicidal ideation or agitation. There was no recording of any cardiac or vascular events within six months of treatment. Six patients (8.6%) self-reported tobacco cessation following varenicline treatment and this was documented for a median of 336 days.
Commentary: We have a group of people where smoking, probably the most important social determinant of health, is almost ubiquitous. The potential side effects of varenicline need to be taken seriously. However, this study starts to suggest that it shouldn't be automatically excluded in this population - it is too easy to discount it from the start and miss the opportunity to help address smoking. One could point to the low cessation rate - just under 9% even with varenicline but, if you want high success rates with any smoking cessation then you're going to be disappointed and these numbers are pretty typical.
A systematic review and meta-analysis of naltrexone implants for the treatment of opioid dependence
Larney S, Gowing L, Mattick RP, et al. Drug Alcohol Rev 2014;33:115-28
This systematic review included studies that compared naltrexone with another intervention or placebo. A total of five randomised trials (n=576) and four non-randomised trials (n=8358) were eligible for inclusion. Naltrexone implants were superior to placebo implants and oral naltrexone in reducing opioid use. There was no difference in opioid use between naltrexone implants and methadone maintenance.
Commentary: The "joy of naltrex". It's difficult to talk about this for too long without getting sucked into an ethical discussion. Clinicians who are fortunate enough to work in societies where the full range of therapeutic options is available will be more cautious. Interestingly the authors use the phrase "suppressing opioid use" when describing the outcome data. It's probably not deliberate but language is important. Suppression as opposed to reduction - and that's the rub with naltrexone implants. They also comment: "the evidence has little clinical utility in settings where effective treatments for opioid dependence are used." The message from this paper is received loud and clear - naltrexone implants should be regarded by any right-minded person as currently being an experimental treatment option that still require proper evaluation.