Should patients undergo surgical intervention as soon as they receive a diagnosis of trigeminal neuralgia (TN), or should they wait until pharmacologic treatment fails? This article describes some of the challenges encountered in surgical trials for TN, and provides some guidance for future trials in this area. One future direction for TN research is to use the innovative “cohort multiple RCT design,” This article describes some of the challenges encountered in surgical trials for TN, and provides some guidance for future trials in this area. One future direction for TN research is to use the innovative “cohort multiple RCT design,” which enables multiple trials to be embedded within a single cohort of patients with TN diagnosis, providing an efficient and effective approach to the testing of multiple interventions for TN with each other and with usual care.
Key points
- •
There is no high-quality comparative effectiveness research for surgery versus pharmacologic management or for different surgical techniques.
- •
High-quality evidence (randomized controlled trials [RCTs]) is required to inform routine decision making for patients with trigeminal neuralgia (TN) and their consultants.
- •
The design and conduct of surgery trials using the standard design has numerous challenges (patient preferences, clinician preferences, clinically meaningful outcome measures, learning curves for surgical techniques, irreversibility of results).
- •
The “cohort multiple RCT” design is an innovative alternative design that provides both long-term observational data and a facility for quick and efficient conduct of multiple trials. Unlike standard trials, patient information and consent replicate that found in routine health care wherever possible.
- •
Embedding multiple trials within a cohort of patients with a diagnosis of TN would enable the quick and efficient identification and recruitment of patients to trials of a variety of interventions, and help provide the information that patients and clinicians require.
Introduction–which trials are needed?
Unusually trigeminal neuralgia (TN), a rare disease, can be managed both medically (pharmacologically) and surgically, and there is some evidence of the importance of psychological therapies. So what trials are needed?
Comparison of Medical Versus Surgical Treatments
Surgical management can yield 100% pain relief for 70% of patients for 10 years. Medical management provides 50% pain relief but becomes less effective over time, and as the dose is raised, results in poorer tolerability. Many of these patients eventually opt for surgery, but the best timing for this is still unknown. Although most patients remain on medical management until it fails, there is evidence that patients prefer surgical management . Zakrzewska and colleagues, reviewing 220 patients who had posterior fossa, found 73% said they would have preferred earlier surgery.
There is also evidence that clinicians/surgeons support early surgery for classic cases of TN and those with positive imaging. Others suggest that surgical treatments should be offered only after patients become refractory to medical management, which is defined by Obermann as failure of 2 drugs. Di Stefano and colleagues in their cohort of 200 patients on medical management suggest that medications remain highly effective and only 7% in their cohort needed surgery.
However, there is no rigorous ( ie, randomised controlled trial [RCT]) evidence to support either an early or delayed surgical management compared with pharmacologic management of TN . The recent Cochrane systematic review on neurosurgical interventions in TN identified just 11 RCTs involving 496 patients ; however, most of these trials were biased. None of the high-quality trials compared different surgical techniques with each other or compared surgery with pharmacologic management. The 3 high-quality RCTs compared different surgical techniques with potentially more refined versions of the same technique. There were no RCTs of microvascular decompression (MVD) (the most invasive procedure and only nondestructive procedure) but observational data suggests that it may have the best long-term outcomes for pain relief.
Given patient and clinician preferences and the lack of evidence to support surgery or pharmacologic management, the most important research question for the TN profession is should patients undergo a surgical intervention as soon as the diagnosis has been made (ie, very early in the course of the condition), or should they wait until the conservative (pharmacologic) treatment has failed? In other words, should they receive surgical treatment that provides something very close to a cure (albeit not necessarily permanent) or remain on medication? If early surgery was comparable to (or better than) medical management, this information would affect how patients viewed their options at the time of diagnosis, and provide more flexibility in the decision-making process in the early stages.
Comparison of the Different Surgical Techniques
There are an emerging number of studies comparing different techniques; however, the interpretation of the results of these studies is hampered by differences in the outcomes used and the short duration of outcomes. Future trials should use the same outcomes and also follow patients for a minimum of 5 years.
New and Comparative Drug Trials
Drug trials in TN are few and far between and most drugs used to date have been established antiepileptics; however, there is now a potential for a new drug with good efficacy and better tolerability to be evaluated. Phase 2 studies have been completed using a novel design of enriched enrollment randomized withdrawal (EERW) design in which patients are initially screened, and then all are put on the active drug for a set period. After this period, only those considered to have been responders are allocated to the randomized part of the trial in which the active drug is compared with a placebo. In this design, there is a set time for the trial but nonresponders are encouraged to drop out.
Moore and colleagues have done a systematic review of all the pain trials using the EERW trial design and suggest that these can play an important role if correctly designed but may be difficult to compare outcomes with classic trials. Comparisons of different drugs are also required and whether single or multiple drugs should be used.
Addition of Psychological Therapies
TN has considerable impact on quality of life and patients live in fear of a recurrence of their pain. One small study (n = 15) has shown that spontaneous pain as opposed to pain evoked by a trigger could be driven by emotional factors. There is anecdotal evidence from surgeons and patients that patients are reluctant to touch their faces after surgical treatments in case they trigger an attack. This behavior is also seen in continuation of medications after surgery, especially after stereotactic radiosurgery surgery (SRS).
In summary, there are a number of research questions in the field of TN that require evidence from well-designed RCTs. This article describes a number of challenges in the design and conduct of trials, with a particular emphasis on surgical trials for TN. It will provide some pointers for future trials.
Introduction–which trials are needed?
Unusually trigeminal neuralgia (TN), a rare disease, can be managed both medically (pharmacologically) and surgically, and there is some evidence of the importance of psychological therapies. So what trials are needed?
Comparison of Medical Versus Surgical Treatments
Surgical management can yield 100% pain relief for 70% of patients for 10 years. Medical management provides 50% pain relief but becomes less effective over time, and as the dose is raised, results in poorer tolerability. Many of these patients eventually opt for surgery, but the best timing for this is still unknown. Although most patients remain on medical management until it fails, there is evidence that patients prefer surgical management . Zakrzewska and colleagues, reviewing 220 patients who had posterior fossa, found 73% said they would have preferred earlier surgery.
There is also evidence that clinicians/surgeons support early surgery for classic cases of TN and those with positive imaging. Others suggest that surgical treatments should be offered only after patients become refractory to medical management, which is defined by Obermann as failure of 2 drugs. Di Stefano and colleagues in their cohort of 200 patients on medical management suggest that medications remain highly effective and only 7% in their cohort needed surgery.
However, there is no rigorous ( ie, randomised controlled trial [RCT]) evidence to support either an early or delayed surgical management compared with pharmacologic management of TN . The recent Cochrane systematic review on neurosurgical interventions in TN identified just 11 RCTs involving 496 patients ; however, most of these trials were biased. None of the high-quality trials compared different surgical techniques with each other or compared surgery with pharmacologic management. The 3 high-quality RCTs compared different surgical techniques with potentially more refined versions of the same technique. There were no RCTs of microvascular decompression (MVD) (the most invasive procedure and only nondestructive procedure) but observational data suggests that it may have the best long-term outcomes for pain relief.
Given patient and clinician preferences and the lack of evidence to support surgery or pharmacologic management, the most important research question for the TN profession is should patients undergo a surgical intervention as soon as the diagnosis has been made (ie, very early in the course of the condition), or should they wait until the conservative (pharmacologic) treatment has failed? In other words, should they receive surgical treatment that provides something very close to a cure (albeit not necessarily permanent) or remain on medication? If early surgery was comparable to (or better than) medical management, this information would affect how patients viewed their options at the time of diagnosis, and provide more flexibility in the decision-making process in the early stages.
Comparison of the Different Surgical Techniques
There are an emerging number of studies comparing different techniques; however, the interpretation of the results of these studies is hampered by differences in the outcomes used and the short duration of outcomes. Future trials should use the same outcomes and also follow patients for a minimum of 5 years.
New and Comparative Drug Trials
Drug trials in TN are few and far between and most drugs used to date have been established antiepileptics; however, there is now a potential for a new drug with good efficacy and better tolerability to be evaluated. Phase 2 studies have been completed using a novel design of enriched enrollment randomized withdrawal (EERW) design in which patients are initially screened, and then all are put on the active drug for a set period. After this period, only those considered to have been responders are allocated to the randomized part of the trial in which the active drug is compared with a placebo. In this design, there is a set time for the trial but nonresponders are encouraged to drop out.
Moore and colleagues have done a systematic review of all the pain trials using the EERW trial design and suggest that these can play an important role if correctly designed but may be difficult to compare outcomes with classic trials. Comparisons of different drugs are also required and whether single or multiple drugs should be used.
Addition of Psychological Therapies
TN has considerable impact on quality of life and patients live in fear of a recurrence of their pain. One small study (n = 15) has shown that spontaneous pain as opposed to pain evoked by a trigger could be driven by emotional factors. There is anecdotal evidence from surgeons and patients that patients are reluctant to touch their faces after surgical treatments in case they trigger an attack. This behavior is also seen in continuation of medications after surgery, especially after stereotactic radiosurgery surgery (SRS).
In summary, there are a number of research questions in the field of TN that require evidence from well-designed RCTs. This article describes a number of challenges in the design and conduct of trials, with a particular emphasis on surgical trials for TN. It will provide some pointers for future trials.
Problems with randomized controlled trials
This section describes the problems with the design, implementation, and interpretation of RCTs of interventions to help patients with their health.
Recruitment
RCTs often have difficulty recruiting sufficient numbers of patients. MacDonald and colleagues found that less than a third of 114 multicenter, publicly funded UK RCTs recruited their original target number of patients within the time originally specified. Failure to recruit to target may have implications for the power and generalizability of trial results. The sample populations often do not contain ethnic minorities or other hard-to-reach groups, for example, elderly individuals, making it difficult to apply to general practice. Trials of medical management in TN are all very small.
Ethical issues
In a systematic review of the literature on barriers to participation in RCTs, Ross and colleagues found that concerns with information and consent were some of the major reasons why both patients and clinicians were unwilling to participate in trials. In routine real-world health care, patients are rarely told of treatments that their clinicians cannot with certainty provide nor are patients told their treatment will be decided by chance. On the other hand, in clinical trials providing this type of “full” information before randomization, is regarded as an ethical requirement. The consequence of this “full” information is that patients worry about the uncertainty of treatment outcome, especially if there is the possibility that they may be allocated to placebo. It is acknowledged that for clinicians there is a potential conflict of interest between what is good for the current patient and what is good for future patients. These issues are nicely demonstrated in the anecdote in Box 1 .
A Canadian surgeon participating in a workshop on designing clinical trials. The Canadian surgeon reported explaining a trial to a potential participant and the fact that there was uncertainty about the best treatment. At the end of the discussion the surgeon asked the patient if he had any questions. “Yes” said the patient, “Can you refer me to a surgeon who does know what is the best treatment for me?”
In a recent phase II trial, patients were reluctant to be recruited, as they had reasonable control and tolerability on their current drugs and were concerned that the new drug for TN would upset this balance (currently unpublished). Moreover, in general practice, patients are often given less information about their treatments than that currently required by some ethics committees who are asked to review intervention trials.
Patient Preferences
Standard “open” (unblinded) pragmatic trials often compare an intervention with treatment as usual. Where the “usual care” on offer is available outside the trial, however, the only incentive for the patient to participate (apart from altruism) is to receive the new intervention. If a patient is allocated to treatment as usual, he or she may withdraw from the trial (attrition bias) or exhibit disappointment bias when reporting outcomes. Patients with rare diseases are more reluctant to take part in trials for this reason. There may therefore be a treatment effect, which results from patient preferences and not from therapeutic efficacy. This is a major problem in TN, where destructive treatments give very different results from nondestructive methods or if compared with medical therapies. As surgery is irreversible, patients may prefer to delay this; yet, when asked specifically about timing of MVD most patients in retrospect said they would have wanted surgery earlier.
Treatment Comparisons
For conditions with many potential treatment options, there are often multiple trials conducted, with each potential treatment being trialed, one at a time, in different populations by different research teams, often with heterogeneous outcomes and heterogeneous trial populations. Thus, when treatments need to be compared, they can only be done by indirect methods. The effectiveness of treatments A versus C can be difficult to evaluate if the only trials of treatments that exist are A versus B and B versus C. Indirect comparisons, in which 2 interventions are compared through their relative effect versus a common comparator, can succeed, but sometimes result in significant discrepancies compared with the results of head-to-head randomized trials. Many competing interventions have thus not been compared or have been compared inaccurately, which is a waste of valuable information and money. This is a major problem in TN, where there are no RCTs of MVD and the RCTs that have been done compare surgical techniques and use varying outcome measures at varying time points. It has therefore been very difficult to compare not just surgical trials but medical trials for the same condition.
Diagnosis
An essential of all trials is an accurate description of the participants by using evidence-based diagnostic criteria, as this will enable clinicians to determine if the patients in the trial are representative of their patients. TN was considered to have very clear diagnostic criteria but it is now emerging that there are several variants and the nomenclature has become confusing with terms such as type 1 and 2 TN or TN with concomitant pain. There has also been a group of conditions known as the trigeminal autonomic cephalalgias, which include 4 different conditions. Two of them, short unilateral neuralgiform headache with conjunctival redness tearing and short unilateral neuralgiform headache with any autonomic symptom, may in fact be yet other variants of TN.
TN and its variants are unusual in that the pain is episodic and there are unpredictable remissions and relapses, which makes it even harder to be sure that the end result is due to the intervention rather than the natural history of the disorder.
Timing
New medications undergo a specific standardized pathway so as to become registered, but this is not the case for surgical interventions. A surgical intervention passes through many phases of innovation and refining and has a tipping point at which the intervention is no longer an innovation but a routine procedure. The tipping point (when equipoise is lost) is extremely variable and cannot be predicted, thus making the accurate timing of RCTs difficult. This has generated what has become known as the Buxton law: “It is always too early [for rigorous evaluation] until suddenly it’s too late.”
Thus, the newest intervention for TN, SRS, was first assessed in an exploratory trial to determine its efficacy and this was done in those patients who would benefit most and by surgeons who had the freedom to develop and refine the intervention. In 2001, an RCT by Flickinger and colleagues of this procedure in a multicenter trial showed that 1 rather than 2 isocenters were sufficient to provide pain relief without sensory loss, one of the first refining studies. This could have been followed by a pragmatic trial that included a very broad population and surgeons with a range of expertise so it represented most closely what occurs in general practice. This approach would have provided information on both the short-term and long-term outcomes of SRS and could have addressed cost-effectiveness and quality-of-life questions if outcomes had been assessed independently. This would have then enabled a standard to be set against which audits could be carried out. Schnurman and Kondziolka have suggested an alternative approach to this problem ( Box 2 ).
Aim:
- •
Use publications to chart progress from innovation to general acceptance.
Method:
- •
Assumes that once there is broad acceptance that an innovation is effective, the next series of papers focus on refining the technique.
- •
The point at which there are more papers on refinement than efficacy or effectiveness becomes the PSA point.
- •
Assess authoring group to see if the procedure was being disseminated and the quality of the publications.
Results:
- •
Refining studies increase efficiency, decrease costs, and may have a moderate effect on outcomes.
- •
Initial efficacy studies have a higher impact on patient care.
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
