Traction

Traction is the act of drawing or pulling and relates to forces applied to the body to stretch a given part or to separate 2 or more parts. Currently, traction is used effectively in treatment of fractures. In physiatric practice, use of traction often is limited to the cervical or lumbar spine with the goal of relieving pain in, or originating from, those areas.^[8]

Since the days of Hippocrates, correction of scoliosis also has involved traction. There are various types of traction currently in clinical use. The most common are mechanical, hydraulic or motorized, manual, and autotraction. Mechanical forms of traction use a hydraulic or motorized pulley system with weights, along with a harness or sling device to attach to the patient’s body. Manual traction involves the therapist using his or her hands on the patient’s body, with the body weight of the therapist providing the tractive force. Autotraction is controlled by the patient pulling on bars or handles at the head of the table, without direct involvement of a therapist. Gravitational traction with a tilt table and underwater variations of traction are also in clinical and home use but are less frequently employed than the other forms described.

Physiologic effects of traction

In the cervical spine, the most reproducible result of traction is elongation. In a classic study, Cyriax reported applying force of 300 pounds manually, with a resultant 1 cm increase in cumulative lumbar spine interspace distance.^[9] Studies have shown that optimum weight for cervical traction to accomplish vertebral separation is 25 pounds. Additionally, 2-20 mm elongation of the cervical spine has been shown to be achievable with 25 or more pounds of tractive force. Studies have demonstrated that anterior intervertebral space shows the most increase in cervical flexion of 30°. Traction in the extended position generally is not recommended, because it is often painful and may increase risk of complications from vertebral basilar insufficiency or spinal instability.

Once friction is overcome in the lumbar spine, the major physiologic effect of traction is elongation. Investigators have reported widening of lumbar interspaces requiring between 70-300 pounds of pull. This widening averaged up to slightly more than 3 mm at one intervertebral level. The length of time that the separation persists remains indeterminate, with studies documenting distraction durations of 10-30 minutes after treatment.

Data on dimensional and pressure changes of lumbar disks caused by traction are not conclusive. Decreases in interdiskal pressure with 50-100 pounds of traction have been documented, but evidence exists that some applications actually cause an increase in interdiskal pressure. Therefore, evidence is inconclusive, with much information favoring at least temporary reduction of the herniated component of an abnormal lumbar disk with concomitant traction.

Some theories on the physiologic effects of traction suggest that stimulation of proprioceptive receptors in the vertebral ligaments and monosegmental muscles may alter or inhibit abnormal neural input from those structures. As with other theories to explain the physiology of traction, there is little to no empirical evidence to fully support it.^[10]

Outcome studies related to traction

Very few scientifically rigorous studies exist that allow the effect of traction to be distinguished from the natural history of pathology (eg, radiculopathy). Criteria have been suggested that would allow the true effects of traction to be delineated. These criteria include (1) randomized controlled trials, (2) blind outcome assessments, (3) equivalent co-interventions, (4) monitored compliance, (5) minimal contamination and attrition, (6) adequate statistical power and description of study design and interventions, and (7) relevant, functionally oriented outcomes. No traction outcome study to date has incorporated these criteria. Despite inadequacies in the literature, randomized, controlled trials that meet some of these criteria do provide some insight into the efficacy of traction as a treatment modality. A review of randomized, controlled trials of traction analyzed English language studies done between 1966 and 2001. The only conclusion that could be drawn, based on this review, wasthatthereexistspoorevidence to support the effectiveness of traction for back pain relief.^[11] A subsequent review, by Graham and colleagues, arrived at 2 clinical conclusions; one conclusion favors the use of intermittent traction over a continuous protocol, and the other does not support the use of continuous traction.^[12] The reviewers felt there was inconclusive evidence overall for either form of traction, based upon the methodologic quality of the numerous studies reviewed.

A systematic literature review by Clarke and colleagues further supported the aforementioned conclusions regarding traction for low back pain.^[10] Through an examination of randomized clinical trials, the authors determined that the evidence did not support the intermittent or continuous use of traction alone to treat low back pain in mixed groups of patients suffering from this condition, whether or not sciatica was present. Owing to inconsistent results and methodologic problems in most of the studies involved, the authors also did not recommend traction for patients with sciatica. Clarke and his coauthors also said that because the available research was insufficient, they could not comment on the use of traction in combination with other therapies.

What can be reasonably derived from these studies is that more work needs to be done to be able to make evidence-based recommendations on the application of traction for back pain.^[13] Additional evidence is also needed to evaluate the optimal type and position of the tractive forces for various clinical conditions, as well as to assess the use of traction as a component of a patient’s treatment, rather than as an isolated modality.

Lumbar traction

The Agency for Health Care Policy and Research (AHCPR) review of the literature on traction resulted in a conclusion that "spinal traction is not recommended in the treatment of acute low back problems." In addition, the 1996 and 1999 guidelines published by the UK Royal College of General Practitioners (RCGP) stated that "there is little evidence to support the continued use of traction in the management of acute low back pain (LBP)." Despite these recommendations, the widespread use of lumbar traction remains relatively high, with up to 20% of patients in the United States and 30% of those with low back pain and sciatica receiving traction as a treatment.

Studies that claim improvement after traction report modest and very short-term improvements, with limited or no improvement in overall function. Additionally, these studies have significant design flaws. While a particular group of patients may benefit from a particular type of traction for either short-term or long-term improvement in functional outcome, the literature currently does not identify this patient population. In addition, it is important to note that although high quality evidence supporting the use of traction for the treatment of low back pain is currently scarce, there is likewise insufficient data in the literature to show that traction is not effective for this problem.^[10]

Cervical traction

Few randomized, controlled trials address patient outcomes after cervical traction. While many studies have produced statistically significant findings, the actual clinical significance of those findings is not clear. Some studies have been published on new protocols for cervical traction, as well as on new devices for traction application. The evidence for the efficacy of these devices and methods appears to need further study before widespread application can be made or recommended.

Techniques for applying traction

Cervical traction generally is accomplished with a free-weight–and–pulley system or an electrical, motorized device. Adequate pull is achieved by using a head or chin sling attached to a system that can provide pull in a cephalad direction. Motorized devices are applied easily but require the patient to be attended. Free-weight–and–pulley systems often are used in the home with 20 or more pounds of water or sand and a pulley system attached to a door. If a tractive force of only 20 pounds is possible, the system is likely to fail to achieve therapeutic results. Advise patients not to attempt cervical traction at home alone, because they may find themselves in uncomfortable positions and may need assistance doffing the traction devices.

Most home traction systems are difficult for patients to set up without assistance. Home cervical traction may cause increase in pain or may fail to produce significant pain relief unless professionally monitored on a periodic basis. At the initiation of home traction, the patient should be required to demonstrate proper use of equipment to the satisfaction of the prescribing physician or therapist.

In the lumbar spine, adequate pull with weights and pulleys or motorized devices to achieve vertebral distraction usually can be obtained with the proper apparatus. Generally, a harness is attached around the pelvis (to deliver a caudal pull), and the upper body is stabilized by a chest harness or voluntary arm force (for the cephalad pull). Motorized units have the advantage of allowing intermittent traction with less practitioner intervention. If the goal of tractive force is to distract lumbar vertebrae, 70-150 pounds of pull usually are needed. Friction between the treatment table and patient's body usually requires tractive force of 26% of the total body weight before effective traction to the lumbar spine is possible. Many traction devices use a split table that eliminates the lower body segment friction.

Body weight theoretically should provide enough pull to distract lumbar vertebrae and eliminate mechanical devices. Gravity traction is applied almost exclusively in the lumbar region. After 10 minutes of inversion traction, documented increases in intervertebral separation are noted; however, side effects also are frequently reported, including increased blood pressure, periorbital petechiae, headaches, blurred vision, and contact lens discomfort.

A study from Hungary re-analyzed an old method of applying traction in the treatment of patients with lumbar or cervical diskopathy.^[14] Patients were vertically suspended by a special harness in a warm-water bath, with a specified amount of weight applied to the lower limbs. One harness allowed for traction on the lumbar spine, while the other focused on the cervical region. The study participants had land-based physical therapy exercises and the weightbath therapy, while a control group only had the exercises. Therapeutic benefit was perceived to be greater by patients treated with a combination of the weightbath and exercise than it was by patients in the control group, according to result following treatment and at 3-month follow-up.

The treatments were well tolerated, and no adverse effects were reported. Although the study concluded that this form of traction treatment "is a relatively straightforward, non-invasive, and low-cost intervention that can be implemented anywhere," further research may be needed to corroborate the findings of this pilot study. Such investigation may need to be supplemented with cost and feasibility data before widespread implementation is initiated.

Other traction technique considerations

In cervical traction, determining sitting versus supine position is based upon the patient's comfort and ability to relax. Maximal distraction generally occurs between 20-30° of flexion without rotation or side bending.

Studies have shown that, in the cervical spine, larger improvement in range of motion (ROM) with less accompanying pain was noted in patients subjected to intermittent traction of 20 pounds peak (10 seconds on, 10 seconds off, for a total of 15 minutes of treatment time) than in patients subjected to 15 minutes of manual or static traction of 25 pounds. Constant cervical distraction forces of 30 pounds generate maximum vertebral separation in 7 seconds or less, and no further separation is gained by applications of up to 60 seconds.

Supine position is chosen most commonly for lumbar traction since the sitting position may result in outcome-limiting discomfort from the harness. Hip flexion of 15-70° routinely is incorporated to cause relative lumbar spine flexion; this may facilitate optimal vertebral separation.

Studies, in addition to patient preference, suggest that some relative advantage exists to an intermittent versus continuous protocol of cervical traction. Some studies report that continuous traction is necessary in the lumbar spine to fatigue muscles and allow strain to fall on joints; however, no statistical difference has been observed with either continuous traction of 100 pounds for 5 minutes or intermittent traction of 100 pounds, peaking for 15 minutes. As in traction on the cervical spine, improved patient tolerance favors an intermittent protocol.

In the sitting position, application of approximately 10 pounds is required to counterbalance the patient's head in cervical traction. Traction of 30 pounds applied to a neck flexed up to 24° can cause vertebral separation, but an increase of force to 50 pounds has been found to produce no clear-cut additional separation. In the lumbar spine, a pull, which equals approximately 50% of the weight of the body part, is needed to overcome friction. As previously noted for the lower body, this amounts to approximately 26% of total body weight.

In 2006, Akbino and colleagues published a study examining what the most beneficial amount of total body weight (TBW) would be for cervical traction.^[15]Trials were done with patients randomly assigned to 1 of 3 groups, with each group receiving traction of 7.5%, 10%, or 15% of the patient’s TBW. The patients in the treatment group using 10% of their TBW demonstrated the highest therapeutic efficacy with the fewest side effects, compared with the 7.5% and 15% TBW groups.

The optimal duration of traction has not been demonstrated clearly. Studies have revealed recommendations varying from 2 minutes to 24 hours in the cervical spine. Duration of approximately 15-25 minutes commonly is prescribed. Cervical traction generally is prescribed at a frequency of daily for the first week and then every other day (ie, 3 times per week) for total treatment duration of approximately 3-4 weeks. In the lumbar spine, treatment generally is recommended in the 8-40 minute range per session, daily for the first week and then every other day (ie, 3 times per week) for a total of 3-4 weeks. In cervical and lumbar traction, goals of treatment determine the time course, as well as the end point of treatment. Possible treatment end points may include pain relief, achievement of normal ROM, return to work or other desired activity, lack of improvement in symptoms, and inability of the patient to cooperate with treatment.

Indications for traction

The literature does not give clear indications what types of neck or low back pain may improve from traction. Studies strongly suggest that traction does not produce significant influence on long-term outcome of neck pain or low back pain. Practitioners who rely on sound scientific advice may use traction rarely. Practitioners who are receptive to empirical treatments may be amenable to the concept that traction may separate vertebrae and decrease the size of herniated disks, thereby benefiting radiculopathy; however, no consensus has been reached among clinicians or researchers in this area.

In a 2008 review investigating the use of lumbar traction for patients with chronic low back pain, Gay and Brault found only 10 randomized, controlled trials addressing this treatment.^[16] As a group, the studies contained more evidence against the use of traction than they did for it. The authors broke the information into subcategories based on whether the data covered patients with back and lower limb pain or with low back pain alone. They also looked at sustained and intermittent traction in these patient groups.

The results indicated a lack of benefit in the use of sustained traction for chronic low back pain, with or without lower limb symptoms. Motorized, intermittent traction, which has been aggressively marketed (eg, VAX-D, DRX9000), likewise did not seem to differ in efficacy from simple intermittent axial traction. Gay and Brault cautioned against fully extrapolating the results of the available randomized, controlled trials to the distraction-manipulation therapies at present, until further research can be completed specifically assessing their effects. At this time, there are at least 2 trials of these devices underway.

Contraindications to traction

No scientific reports clearly delineate contraindications for traction. The practitioner must rely on empirical information and opinion. Old age has been cited as a relative contraindication. Most practitioners agree that contraindications to cervical or lumbar traction include, but may not be limited to, the following: (1) ligamentous instability, (2) osteomyelitis, (3) diskitis, (4) primary or metastatic tumor, (5) spinal cord tumor, (6) severe osteoporosis, (7) clinical signs of myelopathy, (8) severe anxiety, and (9) untreated hypertension.

In the cervical spine, the practitioner also must take into account the fact that patients with vertebral basilar artery insufficiency may be more susceptible to cerebrovascular complications. Furthermore, patients with advanced rheumatoid arthritis or connective tissue disorders may be at risk for atlantoaxial instability.

Other relative contraindications to traction in the cervical or lumbar spine include (1) midline herniated nucleus pulposus, (2) acute torticollis, (3) restrictive lung disease, (4) active peptic ulcer, (5) hernia, (6) aortic aneurysm, and (7) pregnancy.

Akbino and colleagues monitored systolic and diastolic blood pressure, heart rate, rate pressure product, and electrocardiogram (ECG) at 5-, 10-, and 15-minute intervals during administration of cervical traction of 7.5%, 10%, or 15% of the patient’s TBW.^[15] Compared with the patient’s premeasured baseline for these values, there was a decrease in the systolic and diastolic blood pressure and rate pressure product for all subjects in each of the 3 groups. Although the change was not statistically significant in the group receiving traction at 7.5% of TBW, it was (p < 0.05) for the 10% and 15% TBW groups. There was no significant difference in heart rate or ECG variables in any group. The authors of this study recommended monitoring these values before and immediately following application of cervical traction, especially in at-risk patients, or those with known blood pressure or cardiac problems.

Referral considerations

The physiatrist who refers patients for traction must write a detailed and specific prescription that includes at least the following patient information: (1) age, (2) sex, (3) diagnosis, (4) underlying medical conditions, (5) precautions needed, and (6) recommended follow-up. Traction should not be a single treatment approach but rather should be 1 part of a comprehensive rehabilitation treatment program. The most effective use of traction is likely to improve the patient's activity level, mobility, and overall function.

Specific items to outline in traction referrals also should include the following information: (1) position (of the body, neck, or hip and knee), (2) mode of application (continuous or intermittent), (3) weight to be applied, (4) concurrent modalities (eg, heat), (5) frequency and duration of treatment, (6) reevaluation guidelines and time frames, (7) guidelines for discontinuation, and (8) therapeutic goals.

The future of traction

Traction has enjoyed a long history of clinical acceptance based upon very little scientific understanding of its mechanism of action or efficacy. Two 2005 surveys examined the use of traction by physical therapists. One, by Haarte and colleagues^[17] contains data from the United Kingdom, while the second, by Poitras and coauthors,^[18] was performed in Quebec, Canada. Both surveys reported widespread use of traction by physical therapists for the treatment of low back pain. Significant questions exist regarding duration of symptom relief and other benefits derived. Given the difficulty of objective documentation of the benefits of traction, it is not surprising that there has been a reduction in its use.