Federico BalaguƩ; Jean Dudler
Abstract and Introduction
Abstract
This article summarizes the available evidence on the management of
patients with subacute or chronic low back pain. The largest part is
devoted to nonspecific low back pain but the models of spinal stenosis
and disk herniation/sciatica are also specifically addressed. The
authors point out the limited evidence available and the importance of a
tailored approach for the individual patient. As the effect sizes of
most therapies are rather small (close to that of a placebo), patients'
preferences and other variables important for individualized management
are highlighted. The task for the practitioner is difficult and
awareness of this is important. Some speculation regarding potential
future ways of improving patient care are presented.
Introduction
Low back pain (LBP) remains the most frequent musculoskeletal
complaint worldwide and all age groups are affected by these symptoms.
They are classically stratified into acute, subacute and chronic, with
respective cut-offs of <6 weeks, 6–12 weeks and >12 weeks.
By itself, it produces direct and indirect costs of hundreds of
billions of dollars for the US alone. Recent studies in adults and
elderly populations have shown a significant increase in LBP, both in
numbers and costs, in terms of investigations, treatments and
disability, an observation at least partially explained by a raise in
prevalence.
However, the large differences in the rate of spinal surgical procedures observed between states within the US,
as well as between countries worldwide, suggest that decision-making is
certainly influenced by regulations and other sociopolitical factors.
As LBP is extremely prevalent, the main problem remains the chronic
cases, in particular in term of investigations and costs. Acute episodes
of LBP statistically have quite a good prognosis more or less
independently of the chosen treatment. A recent review confirms that a
variety of treatments of acute LBP are effective and supported by the
literature. Moreover, there are excellent updated reviews on the management of acute pain not limited to but including LBP.
The interested reader can download this electronically.
If until recently a figure of 8–10% was usually accepted as the
number of acute LBP episodes evolving into chronic cases, recent studies
have show much more ominous figures with frequent relapses and
persistence of symptoms at 1 year in up to 10–30% of cases according to
definitions used. On the other hand, more than a third of the patients
with LBP for more than 3 months do recover within 12 months.
Defining if a patient is going to become chronic or establishing an
individual prognosis based on epidemiological studies is a very
difficult task. Certainly, a precise diagnosis would help. However, it
is commonly accepted that a specific identifiable etiology is only found
in around 15% of cases, including disk herniations, spinal stenosis,
osteoporotic fractures, inflammatory diseases and the infrequent
(approximately 1%) specific neoplastic or infectious destructive
lesions.
The largest part of this manuscript is devoted to the 85% of patients
asking for medical attention and suffering from chronic LBP without any
of those specific identifiable etiologies, the so-called nonspecific
(NS) LBP. Furthermore, we included spinal stenosis and lumbar disc
herniation in the discussion in regard to their frequency in daily
practice.
It has been shown already in adolescent populations that psychosocial
factors are stronger predictors of incident LBP than mechanical
factors.
In adult populations, psychosocial factors are risk factors for
chronicity much more strongly related to outcome than any clinical or
mechanical variables,
while previous episodes of pain are strong predictors of future ones.
Twin's cohort studies have shown that NS-LBP is >40% genetically
determined, whilst work, leisure time and physical activities play a
minor role.
If the natural history of acute episodes of LBP is favorable
independently of the chosen treatment, our daily concern remains chronic
LBP and we have focused this review on the management of chronic and
subacute LBP cases.
Finally, there is an overwhelming amount of literature on the
subject, as well as numerous guidelines and recommendations. This short
overview is based, for practical reasons, on the latest guideline we
were aware of,
an English study with several major strengths, such as including among
their criteria for implementation the likelihood of impact on patients'
outcome and efficient use of NHS resources, completed with relevant
randomized controlled trials (RCT) or meta-analysis published more
recently.
Management of LBP
All of us, including patients, would prefer to prevent rather than to
treat. Primary prevention would ideally prevent the occurrence of LBP,
while secondary preventive measures are aimed at preventing the
recurrence of acute LBP episodes with their risk of chronification,
which is the most relevant problem.
Primary Prevention
During the last few decades it has been shown that adolescents report
NS spine pains with a frequency close to that of their adult
counterparts. These figures indirectly preclude any major efficacy of
primary prevention techniques and suggest that any preventive measures
should be implemented very early in an individual's life to have any
chance to prevent the occurrence of LBP.
Nevertheless, numerous interventions have been tested over the years,
and the evidence available for primary prevention of back problems have
been recently reviewed.
Only exercise interventions, without any specificity of type, have
shown effectiveness using the highest quality criteria, with an effect
size (ES) ranging from 0.39 to >0.69 (ES computation: ES = [mean1 -
mean2]/[pooled SD]; with ES interpretation: <0.15 = negligible
effect; >0.15 and <0.40 = small effect; >0.40 and <0.75 =
medium effect; and >0.75 = large effect).
Other techniques such as
stress management, shoe inserts, back supports, ergonomic/back education
and reduced lifting programs have not been found to be effective.
Secondary Prevention
As stated before, an acute episode of LBP has an intrinsically good
prognosis more or less independent of the chosen treatment. A variety of
treatments are efficient for the acute episode,
but the question is if any early intervention in this setting could or
would prevent the ominous chronification and persistence of the problem
in a significant percentage of patients. Up to 30% of cases will evolve
badly, but there is no specific validated therapy of an isolated acute
phase that would prevent this evolution and work in secondary prevention
of chronic NS-LBP, except the physical exercises previously recommended
as primary prevention.
Conservative Treatment of Chronic NS-LBP
As no preventive measure has sufficient power to prevent chronic
NS-LBP, we are left with managing the problem when it arises, which can
be done with an array of approaches, from conservative therapy to
surgical intervention. In 2009, Rainville
et al. reported on
the evidence on conservative treatments for chronic LBP looking at the
nonsurgical arm of several RCTs comparing surgical and conservative
management.
Clearly, surgery primarily focused on the alteration of structures
perceived to be the sources of pain whilst conservative management aims
to improve patients' function, with or without simultaneous improvement
of pain.
The poor results obtained in terms of public health are not due to a
lack of therapeutic possibilities. In an amazing paper, Haldemann
reported that nonexhaustive research identified more than 200 treatments
for LBP.
In fact, while lack of treatments is not a problem, overtreatment could
be a more worrisome problem, and opinion leaders have even suggested
that clinicians back off.
Finally, lack of a really efficient and universal therapy remains a
problem. A review of the magnitude of the effect of different treatments
in acute and chronic LBP shows that the average effects of treatments
for NS-LBP are not much greater than those of placebos.
For example, NSAIDs and muscle relaxants reduce the intensity of pain
by less than 20 points on a 100-point scale both for acute and chronic
LBP patients. The very few therapies that have demonstrated larger
effect sizes (>30 on a 0–100 pain scale) have only been evaluated in
single small studies, and not been reproduced in any larger cohort.
The overwhelming number of available guidelines and recommendations
reflects the difficulty of managing a common problem in the absence of
any universal efficient treatment. We chose to highlight and comment on
the latest UK guidelines for NS-LBP between 6 weeks and 12 months
duration,
which have the advantage of not only summarizing the main
recommendations for patients in seven headings (reused as subheadings
below), but also to clearly define therapeutic modalities that should
not be prescribed despite the urge to be proactive in front of a
suffering patient.
Information, Education & Patient Preferences
Promoting self-management and encouraging physically activity
certainly make sense and should be reasonably cheap. The question is how
much resource should be invested in this direction. The limits of
education and self-management have recently been highlighted for
osteoarthritis,
and the same caveat certainly applies for LBP.
Along the same lines, offering booklets and stand-alone formal
education programs could appear appealing, as they are widely available.
However, there is no scientific evidence, and it seems essential to
take into account the person's expectations and preferences before using
such programs.
Physical Activity & Exercise
Again, advising people with LBP to stay physically active is likely
to be beneficial, and advising to exercise is adequate. Nevertheless,
advising is only part of the problem. There are often a lot of concerns
from practitioners to know which type of exercise program should be
ideally prescribed, but most types of exercise will be appropriate,
including aerobic activity, movement instruction, muscle strengthening,
postural control or stretching. The real trick is actually to motivate
the patient to exercise, and a structured group exercise program is the
recommended first step. A one-to-one supervised exercise program may be
offered if a group program appears unsuitable for a particular person,
and is certainly more adequate than leaving the patient to exercise on
their own, regardless of their good resolutions.
Van Midelkoop
et al. have recently summarized the evidence
for exercises. "Exercise therapy seems to be effective for the
prevention of LBP, but only a few recent trials have been conducted.
This therapy is not effective for acute LBP, whereas it is effective for
chronic LBP; however, there is no evidence that any type of exercise is
clearly more effective than others. Subgroups of patients with LBP
might respond differently to various types of exercise therapy, but it
is still unclear which patients benefit most from what type of exercise.
Adherence to exercise prescription is usually poor, so supervision by a
therapist is recommended. If home exercises are prescribed, strategies
to improve adherence should be used. Patient's preferences and
expectations should be considered when deciding which type of exercise
to choose".
In other words, one can prescribe exercise therapy without the fear of
not being a specialist as there is no clear cut benefits for one type to
the other, or rather we are still unable to precise who is going to
benefit from exercises. Again, the most important and hardest part is
getting the patients' adherence to the program; matching the patients'
expectations and preferences should help.
Manual Therapy
The UK guideline proposes to consider offering a course of manual
therapy, including spinal manipulation. However, not all patients feel
comfortable with this type of approach, and, again, patient's
expectation and preferences clearly dictate the use of such therapy.
Other Nonpharmacological Therapies
We can only agree with the guideline's authors in their strong
recommendation not to offer any of the multiple therapies with no
scientific support, including laser therapy, interferential therapy,
therapeutic ultrasound, transcutaneous electrical nerve stimulation
(TENS), traction or lumbar supports.
Invasive Procedures
Injections and denervations are other fashionable procedures that
have gained large acceptance in some countries. However, a systematic
review on injection therapy and denervation procedures for chronic LBP
has recently concluded that the evidence supporting these two categories
of therapies over placebo is "low to very low quality". The authors
highlight that it cannot be ruled out that in carefully selected
patients some injection therapy or denervation procedures may be of some
benefit;
however, it remains equally false to push those procedures for the
majority of patients and the British guidelines recommend not to offer
injections of therapeutic substances into the back for NS-LBP.
Acupuncture is a special case that could be considered for a limited
number of sessions, and is probably more beneficial if it matches the
patient preferences.
Combined Physical & Psychological Treatment Program
Combined physical and psychological treatments (including cognitive
behavioral approach and exercise) have been shown to be efficient.
However, to demonstrate benefits they must be quite substantial,
comprising around 100 h over a maximum of 8 weeks. Availability of such
programs and of the patient are limiting factors, but cost issues remain
the main limitation and such programs should be reserved for patients
with high disability and/or significant psychological distress and who
have failed at least one less intensive treatment program.
Group cognitive behavioral treatment has been shown to have a
statistically significant effect (over 1 year) at much lower cost on
troublesome subacute and chronic LBP in primary care,
with effect sizes ranging from 0.1 for SF-12 mental to 0.5 for SF-12
physical and fear-avoidance beliefs. However, the benefits appear
limited and are also clearly dependent on local availability of such
programs.
Pharmacological Therapies
As in all pain-related guidelines, regular paracetamol is the first
recommended medication option. However, paracetamol is not free of side
effects when taken regularly at a recommended dose. NSAIDs and/or weak
opioids are the next step, again despite the fact that their benefits
are far from being established.
NSAIDs are also far from being side effect free, particularly in the
elderly. It is important to take into account the individual risk, and
in particular the gastrointestinal risk, and either a standard NSAID
coprescribed with a proton pump inhibitor (PPI) or a COX-2 inhibitor is
recommended. Again, the patient's profile, preferences and expectations
should not be forgotten. Aspirin cancels the benefits of COX-2
inhibitors, while more than 25% of the patients never start their PPI
cotherapy. In other words, we often take considerable risk for a therapy with limited evidence for efficacy.
If ineffective, recommendations consider offering tricyclic
antidepressants for pain relief. However, these are not more efficient
than the other analgesics discussed above. Selective serotonin reuptake
inhibitors (SSRIs) are usually not proposed for treating pain,
but a recent RCT on the efficacy of duloxetine in patients with
non-neuropathic chronic LBP has shown a significant reduction in pain
and improved function compared with placebo.
Finally, one can consider offering strong opioids for short-term use
to people in severe pain. Referral for specialist assessment may be
required for prolonged use of strong opioids given the risk of opioid
dependency and side effects. There is also increasing concern about the
utilization of opioids for chronic noncancer pain management.The adverse effects
and the utilization of these drugs in rheumatology
have recently been reviewed. A recent Cochrane review, including, among
others, seven studies on LBP patients, highlights the limits of the
tolerance and efficacy of these drugs.
While opioids are an alternative, they are not magical pills that will solve the problem of pain management in LBP.
If no treatment is universally and totally efficient, it certainly
appears rational to combine different interventions, a commonly used
practice for some LBP healthcare providers. A recent Cochrane review on
combined chiropractic interventions reported that combined interventions
slightly improved pain and disability in the short term and pain in the
medium term, but only for acute and subacute LBP. No difference was
demonstrated for chronic LBP and for studies including a mixed
population of LBP.
Even if combining several treatments improves the results, that
approach is not always cost effective, as recently shown by Smeets
et al.
There is an urge to be proactive and we often use and abuse unproven
therapeutics. However, we should at least base our decisions to continue
such treatments on the individual response.
Conclusion
The societal burden of LBP keeps increasing despite, or perhaps
because of, the ever increasing number of diagnostic and therapeutic
procedures performed for this very common ailment. Happily, the natural
history of acute episodes of LBP remains favorable in most cases,
independently of the chosen treatment. Subacute and chronic cases
represent the real challenge and our daily concern.
We are still unable
to adequately identify the patients at high risk of becoming chronic,
nor has any universal measure been demonstrated useful for primary or
secondary prevention. Furthermore, overtreating patients with NS-LBP is
probably more deleterious than beneficial and we should probably
restrain from being overenthusiastic at using one of the hundreds of
treatments described for the management of NS-LBP at the first sign of
LBP.
Finally, the risk of potential side effects should also be weighted
in the balance, as well as the individual patient's preferences taken
into account, before starting any therapy. We should ensure that we have
identified the reasons why the patient is sitting in front of us,
bearing in mind that among individuals reporting LBP, "consulters" and
"nonconsulters" cannot be distinguished in terms of pain intensity.We still misunderstand too often the motivation and/or expectations of
the individual patient, a problem coupled with the limited knowledge of
the psychological profile, patient preferences, CNS participation, and
so on, based on the meager time available for a clinical appointment.
There is limited evidence for a majority of treatments in chronic
LBP, and effect sizes are usually moderate for the few statistically
significantly effective forms of treatment. We are also faced with
difficulties in interpreting the evidence, as review articles may end up
with significantly different conclusions based on the same literature,
and the difficulties in using evidence in clinical practice have been recently highlighted.
However, the individual response cannot always be inferred from the
limited evidence available, and patients should still be managed despite
the absence of universally efficient treatment. We apply the same
treatments with their limited evidence and small effect sizes to all
chronic LBP patients. More precise diagnosis and subgrouping of NS-LBP
for the purposes of treatment might improve the efficacy of therapies;
however, a recent review of the topic has concluded: "At this point,
the bulk of research evidence in defining subgroups of patients with LBP
is in the hypothesis generation stage; no classification system is
supported by sufficient evidence to recommend implementation into
clinical practice".
Spinal stenosis and disk herniation with sciatica are good examples
that our subgrouping is still too vague to be really useful.
We should promote exercise and self-management programs for osteoarthritis and back pain. Despite weak evidence for chronic back pain, exercise programs appear to represent the best way forward,
and there is also moderate-quality evidence that post-treatment exercise programs can prevent recurrences of back pain.
While patients' self-management and the promotion and encouragement
of the maintenance of daily physical activities can, and certainly
should, be encouraged in all patients at no risk and no cost, there are a
multiplicity of treatments where the risk/cost– benefit ratio is not so
clear.
Even simple measures such as the prescription of analgesics or
NSAIDS should be monitored by means of validated tools in order to
evaluate the outcome. In the absence of clear and established benefits
for any therapy, it is essential that any prescribed treatment is
evaluated and monitored at the individual level. More difficult with the
urge to be proactive in front of a suffering patient, it is mandatory,
particularly in a time of limited healthcare resources, to refrain from
using all those therapies and procedures where clear lack of benefits
has been demonstrated.
It is possible that some of those therapies remain valid for some
individual patients or well-defined subgroups of LBP. However, so far we
have been unable to identify and characterize such subgroup well enough
to be applicable at the individual level. The concept of personalized
and individualized healthcare should not be used to promote the use of
inadequate therapies, whose evaluation in such settings should be
clearly limited to well-designed trials.
More than anything, we should try to demedicalize LBP and promote
self-management as much as possible. Promoting exercises with methods
that do not require any contacts with healthcare providers, like
walking, may be effective for the treatment of LBP (low-to-moderate
evidence in a recent review),
and as recently writen by Weiner and Nordin, "a large proportion of
patients seeking care can manage their short term and even longer term
incapacity".
It has been shown that acceptance of pain is significantly associated with quality of life.
[62]
We still do not know to what extend this variable can be influenced by
the healthcare providers, but it is all too easy to lure patients into
hopes that specific diagnosis and miracle treatments are available.