Tipos de fixação de Ligamentoplastias LCA / Graft Selection in ACL reconstruction

Graft Selection in ACL reconstruction



History

The type of graft that the surgeon chooses for ACL reconstruction has evolved over the past few decades. In the 1970s, Erickson popularized the patellar tendon graft autograft that Jones had originally described in 1960. This became the most popular graft choice for the next three decades. In fact, in a survey of American Academy of Orthopaedic Surgeon members done in 2000, 80% still favored the use of the patellar tendon graft.

In the light of harvest site morbidity and postoperative stiffness associated with the patellar tendon graft, many surgeons began to look at other choices, such as semitendinosus grafts, allografts, and synthetic grafts. Fowler and then Rosenberg popularized the use of the semitendinosus. However, even Fowler was not convinced of the strength of the graft. Then, Kennedy and Fowler developed the ligament augmentation device (LAD) to supplement the semitendinosus graft. Gore-Tex (Flagstaff,AZ), Leeds-Keio, and Dacron (Stryker, Kalamazoo, MI) were choices for an alternative synthetic graft to try to avoid the morbidity of the patellar tendon graft. The initial experience was usually satisfactory, but the results gradually deteriorated with longer follow-up.

Allograft was another choice that avoided the problem of harvest site morbidity. The initial allograft that was sterilized with ethylene oxide had very poor results. Today the freeze-dried, fresh-frozen, and cryopreserved are the most popular methods of preservation of allografts.

The allograft has become a popular alternative to the autograft because it reduces the harvest site morbidity and operative time. However, Noyes has reported a 33% failure with the use of allografts for revision ACL reconstruction.

The aggressive postoperative rehabilitation program advocated by Shelbourne in the 1990s greatly diminished the problems associated with the patellar tendon graft. Before that, the patient had to be an athlete just to survive the operation and rehabilitation program. Theaggressive program emphasized no immobilization, early weight bearing, and extension exercises.

There was renewed interest in the semitendinosus during the mid-1990s. Biomechanical testing on the multiple-bundle semitendinosus and gracilis grafts demonstrated them to be stronger and stiffer than other options. This knowledge combined with improved fixation devices such as the Endo-button gave surgeons more confidence with no-bone, soft tissue grafts. The Endo-button made the procedure endoscopic, thereby eliminating the need for the second incision. Fulkerson, Staubli, and others popularized the use of the quadriceps tendon graft. This again reduced the harvest morbidity, especially when only the tendon portion was harvested.

Shelbourne has described the use of the patellar tendon autograft from the opposite knee. He claims that this divides the rehabilitation between two knees and reduces the recovery time. With the contralateral harvest technique, the average return to sports for his patients was four months. With both the patellar tendon and the semitendinosus added to the list of graft choices, the need for the use of an allograft is minimized.

The latest evolution is to use an interference fit screw to fixate the graft at the tunnel entrance. This produces a graft construct that is strong, short, and stiff. It means that the surgeon now has to learn just one technique for drilling the tunnels and can chose whatever graft he or she wishes: hamstring, patellar tendon, quadriceps tendon, or allograft.

Successful ACL reconstruction depends on a number of factors, including patient selection, surgical technique, postoperative rehabilitation, and associated secondary restraint ligamentous instability. Errors in graft selection, tunnel placement, tensioning, or fixation methods may also lead to graft failure. Comparative studies in the literature show that the outcome is almost the same regardless of the graft choice. The only significant fact from the metaanalysis, as confirmed by Yunes, is that the patellar tendon group had an 18% higher rate of return to sports at the same level. The most important aspect of the operation is to place the tunnels in the correct position. The choice of graft is really incidental. Studies by Aligetti, Marder, and O’Neill show that the only significant differences among the grafts is that the patellar tendon graft has more postoperative kneeling pain.

Evolution in Graft Choice at Carleton Sports


Medicine Clinic

The most popular graft in the early 1990s was the patellar tendon graft

(Fig. 1). With the evolution of the 4-bundle graft and improved fixation in the mid-1990s, the hamstring graft became more popular. The swing to hamstring grafts then became largely patient driven.When the patients went to therapy after the initial ACL injury, they saw how easy the rehabilitation was for the hamstring tendon and opted for that graft.

The main choices of graft for ACL reconstruction are the patellar tendon autograft, the semitendinosus autograft, and the central quadriceps tendon, allograft of patellar tendon, Achilles tendon, or tibialis anterior tendon, and the synthetic graft.

Figure 1. The evolution of the graft choice. The white bar is the hamstring graft.


Patellar Tendon Graft

The patellar tendon graft was originally described as the gold-standard graft. It is still the most widely used ACL replacement graft (i.e., it is used in approximately 80% of cases), but it is not without problems. Shelbourne has pushed the envelope further with the patellar tendon graft. He has recently reported on the harvest of the patellar tendon graft from the opposite knee, with an average return to play of four months postoperative.

The advantages of the patellar tendon graft are early bone-to-bone healing at six weeks, consistent size and shape of the graft, and ease of Patellar Tendon Graft harvest. The disadvantages are the harvest site morbidity of patellar tendonitis, anterior knee pain, patellofemoral joint tightness with late chondromalacia, late patella fracture, late patellar tendon rupture, loss of range of motion, and injury to the infrapatellar branch of the saphenous nerve. Most of the complications are the result of the harvest of the patellar tendon.This is still the main drawback to the use of the graft.

Patellar Tendon Graft Indications

The ideal patient for an ACL reconstruction is the young, elite, competitive, pivotal athlete. This is the young athlete who wants to return to sports quickly and is going to be more aggressive in contact sports for a longer period of time. There is no upper age limit for patellar tendon reconstruction, but the younger athlete has more time to commit to knee rehabilitation. If the patellar tendon is the gold standard of grafts, then this is the graft of choice for the professional, or elite, athlete. Finally, the competitive athlete understands the value of the rehabilitation program and will not hesitate to spend three hours a day in the gym. The author’s assessment is that 50% of the success is the operation, and 50% is the rehabilitation program.

Pivoting Activities

The ACL is only required for pivotal athletics. Most nonpivotal athletes can usually cope without an ACL. Cyclists, runners, swimmers, canoeists, and kayakers, for example, can function well in their chosen sport without an intact ACL.

Athletic Lifestyle

This operation should be reserved for the athletic individual. In most activities of daily living the ACL is not essential. If the nonathlete has giving way symptoms, it is probably the result of a torn meniscus and not a torn ACL.The meniscal pathology can be treated arthroscopically, and the patient can continue with the use of a brace as necessary.

Patellar Autograft Disadvantages

Harvest Site Morbidity

The main disadvantage of the patellar tendon graft is the harvest site morbidity. The problems produced by the harvest are patellar tendonitis, quadriceps weakness, persistent tendon defect, patellar fracture, patellar tendon rupture, patellofemoral pain syndrome, patellar entrap-ment, and arthrofibrosis. The common long-term problem is kneeling pain.

Kneeling Pain

The most common complaint after patellar tendon harvest is kneeling pain. This can be reduced by harvesting through two transverse incisions. This reduces the injury to the infrapatellar branch of the saphenous nerve.

Patellar Tendonitis

Pain at the harvest site will interfere with the rehabilitation program. The strength program may have to be delayed until this settles. The problem is usually resolved in the first year, but it can prevent some high performance athletes from resuming their sport in that first year.

Quadriceps Weakness

The quads weakness may be the result of pain and the inability to participate in a strength program. If significant patellofemoral symptoms develop, the athlete may be unable to exercise the quads.

Persistent Tendon Defect

If the defect is not closed, there may be a persistent defect in the patellar tendon. This results in a weaker tendon.

Patella Entrapment

If the defect is closed too tight, the patella may be entrapped, and patellar infera may result. This will certainly result in patellofemoral pain, because of an increase in patellofemoral joint compression.

Patella Fracture

The fracture may occur during the operation or in the early postoperative period. Intraoperative patella fracture may be the result of the use of osteotomes. If the saw cuts are only 8-mm deep and 25-mm long, and the base is flat to avoid the deep V cut, an intraoperative fracture is rare. The late fractures are produced by the overruns of the saw cuts. The overruns may be prevented by cutting the proximal end in a boat shape.

The left X-ray shows a displaced transverse patellar fracture, at three months postoperative. The right X-ray shows the postoperative internal fixation with cannulated AO screws and figureof-eight wire.

 

Figure 2. X-ray of displaced transverse patellar fracture at three months postoperative.

Figure 3. X-ray of postoperative internal fixation with cannulated AO screws and figure-of-eight wire

Tendon Rupture

This may occur if a very large graft is taken from a small tendon. The standard is a 10-mm graft, measured with a double-bladed knife. The bone blocks are trimmed to 9 mm to make the graft passage easier.

Patellofemoral Pain

This topic is controversial in the literature. The older literature reported a high incidence of patellofemoral pain associated with ACL reconstruction. However, most of the disability could be blamed on rehabilitation programs that consisted of immobilization.There is no doubt that some patients will develop pain, some will develop crepitus, and some will have tendonitis, but results have improved with more aggressive rehabilitation programs with early motion and weight bearing. To prevent the patella from being bound down, the patella should be mobilized daily by the physiotherapist.

Arthrofibrosis

This severe problem is rarely seen now in ACL reconstructions.The true condition is idiopathic and is probably the result of fibroblastic proliferation. As a result, very little can be done to prevent it. It may be more common in the patient who forms keloid. The more common condition of loss of range of motion may be the result of incorrect tunnel placement or postoperative immobilization. In the mid-1980s, a limited range of motion hinge cast (preventing 30° of extension) was used for six weeks postoperatively, thereby causing problems in regaining extension. Many of these cases required arthroscopic debridement (10–18%, in the first year). The loss of extension was almost completely eliminated by changing to an extension splint. The acceptance of aggressive physiotherapy to regain extension eliminated the problem. This problem of postoperative stiffness made the use of a synthetic ligament, with no immobilization, very attractive. The reoperation for loss of range of motion is now very uncommon.
 
Tendon Rupture


This may occur if a very large graft is taken from a small tendon. The standard is a 10-mm graft, measured with a double-bladed knife. The bone blocks are trimmed to 9 mm to make the graft passage easier.

Patellofemoral Pain

This topic is controversial in the literature. The older literature reported a high incidence of patellofemoral pain associated with ACL reconstruction. However, most of the disability could be blamed on rehabilitation programs that consisted of immobilization.There is no doubt that some patients will develop pain, some will develop crepitus, and some will have tendonitis, but results have improved with more aggressive rehabilitation programs with early motion and weight bearing. To prevent the patella from being bound down, the patella should be mobilized daily by the physiotherapist.

Arthrofibrosis

This severe problem is rarely seen now in ACL reconstructions.The true condition is idiopathic and is probably the result of fibroblastic proliferation. As a result, very little can be done to prevent it. It may be more common in the patient who forms keloid. The more common condition of loss of range of motion may be the result of incorrect tunnel placement or postoperative immobilization. In the mid-1980s, a limited range of motion hinge cast (preventing 30° of extension) was used for six weeks postoperatively, thereby causing problems in regaining extension. Many of these cases required arthroscopic debridement (10–18%, in the first year). The loss of extension was almost completely eliminated by changing to an extension splint. The acceptance of aggressive physiotherapy to regain extension eliminated the problem. This problem of postoperative stiffness made the use of a synthetic ligament, with no immobilization, very attractive. The reoperation for loss of range of motion is now very uncommon..

Contraindications to Harvest of the Patellar Tendon

Preexisting Patellofemoral Pain

Is preexisting patellofemoral pain a contraindication to harvesting the patellar tendon? The conventional wisdom is yes; it would not be a wise procedure in this situation. Rather, it is like hitting a sore thumb with a hammer! In the past, when chondromalacia was seen at the time of arthroscopy, the graft choice would be changed to hamstrings.

The Small Patellar Tendon


The harvesting of the central third of the patellar tendon in a small tendon is more theoretical than practical. The advice in a small patient with a tendon width of only 25 mm would be to take a narrower graft of 8 to 9 mm or use another graft source.


Preexisting Osgoode-Schlatters Disease Shelbourne has reported that a bony ossicle from Osgoode-Schlatters disease is not a contraindication to harvest of the patellar tendon.

Because the fragment usually lies within the bony tunnel, this bone may be incorporated into the tendon graft.

Hamstring Grafts

Advantages of Hamstring Grafts

The main advantage of the hamstring graft is the low incidence of harvest site morbidity. After the harvest, the tendon has been shown by MRI to regenerate. The 4-bundle graft is usually 8mm in diameter, which is a larger cross-sectional area than the patellar tendon.

Disadvantages of Hamstring Grafts

The disadvantage of any autograft is the removal of a normal tissue to reconstruct the ACL. The harvest of the semitendinosus seems to leave the patient with minimal flexion weakness. One study did show some weakness of internal rotation of the tibia after hamstring harvest.

Injury to the saphenous nerve is rare and can be avoided with careful technique. The fixation of the graft remains one of the controversial issues.




Issues in Hamstring Grafts



The major issues with the use of hamstring grafts are:

Graft strength.

Graft fixation.

Graft healing.

Donor site morbidity.

Early rehabilitation.

Graft strength and stiffness.



In one of the earlier studies, Noyes reported that one strand of the semi-t was only 70% the strength of the ACL. However, hecompared this to a 15-mm-wide patellar tendon graft that was 125% the strength of the native ACL. This was widely quoted as a reason to use the patellar tendon graft rather than the hamstring.With the advent of the multiple bundles of hamstrings, this graft now has twice the strength of the native ACL (Fig. 7). Sepaga later reported that the semitendinosus and gracilis composite graft is equal to an 11-mm patellar tendon graft. Marder and Larson felt that if all the bundles are equally tensioned, the double-looped semi-t and gracilis is 250% the strength of the normal ACL. Hamner, however, emphasized that the strength is only additive if the bundles are equally tensioned.



Soft Tissue Fixation Techniques

There are various techniques for securing the soft tissue to the bony tunnel in ACL reconstruction. Each one has strengths and weaknesses. Pinczewski pioneered the use of the RCI interference fit metal screw for soft tissue fixation. The use of a similar type of bioabsorbable screw that was used in bone tendon bone fixation was a natural evolution. To overcome the weak fixation in poor quality bone, the use of a round pearl, made of PLLA or bone, was developed.This improved the pullout strength by 50%.The Endo-button, popularized by Tom Rosenberg, was improved with the use of a continuous polyester tape. This made the fixation stronger and avoided the problems of tying a secure knot in the tape. The cross-pin fixation has proven to be the strongest, but has a significant fiddle factor to loop the tendons around the post. The Arthrex technique is the easiest to use.Weiler, Caborn, and colleagues have summarized the current concepts of soft tissue fixation. The estimates of the force on the normal ACL during activities of daily living are as follows:



Level walking: 169N

Ascending stairs: 67N

Descending stairs: 445N

Ascending ramp: 27N

Descending ramp: 93N



It is commonly quoted that a person needs more than 445N pullout strength of the device just to handle the activities of daily living. However, Shelbourne has reported good results with the patellar tendon graft fixed by tying the leader sutures over periosteal buttons (Ethicon, J&J, Boston, MA). This form of fixation has a low failure strength, but is clinically successful. The gold standard of the interference fit screw fixation of the bone tendon bone, 350 to 750N, has been used to compare the soft tissue fixation.

The pullout strengths also vary from tibia to the femur. The femoral pullout is higher because the tunnel is angled to the graft and the pull is against the screw that is placed endoscopically. In the tibial tunnel, the graft pulls away from the screw in the direct line of the tunnel.

The initial fixation points were at a distance from the normal anatomical fixation of the ACL. The trend has been to move the fixation closer to the internal aperture of the tunnel. This shortening of the intraarticular length has improved the stiffness of the graft.

The pullout strength of bioabsorbable screw can vary widely depending on its composition. The screw fixation has also been shown to be bone quality dependent. These considerations should be taken into account when choosing a femoral fixation device for soft tissue grafts.

Disadvantages

The disadvantages of the hamstring graft are the various methods used to fix the graft to bone, including staples, Endo-button, and interference fit screws. Furthermore, the graft harvest can be difficult, the tendons can be cut off short, and there is a longer time for graft healing to bone, approximately 10 to 12 weeks.


Pullout Strengths of Soft Tissue Devices

The fixation of the graft depends on both the tibial and femoral fixation. The rehabilitation protocol should reflect the type of fixation used. All the femoral fixation devices provide reasonable fixation. The cyclic load is more important than the ultimate load to failure. The interference screw fares worst with cyclic loads.



Interference Fit Screws

The interference fit screw is shown is Figure 4.

Figure 4. The interference screw fixation of the soft tissue graft in a cadaver model

.

Advantages

The advantages are as follows:

Quick, familiar, and easy to use.
Direct bone to tendon healing, with Sharpey’s fibers at the tunnel aperture.
Less tunnel enlargement.

Disadvantages

The disadvantages are as follows:

Longer graft preparation time.
Bone quality dependent.
Damage to the graft with the screw.
Divergent screw has poor fixation.
Removal of metal screw makes revision difficult.

Several refinements have been made to the interference screw technique to increase the pullout strength and cyclic load performance. The end of the graft may be backed up with a round ball of PLLA, the Endo- Pearl (Linvatec, Largo, FL) or bone to abut against the screw and prevent the slippage of the graft under the screw. The tunnels may be dilated or compacted when the bone is osteopenic.A longer screw with a heavy whipstitch in the graft improves pullout strength. The leader sutures from the graft may be tied over a button or post on the tibial side to back up the screw fixation.




Cross-Pin Fixation

The cross-pin fixation is shown in Figure 5

Figure.5. The Arthrex transfix pin fixation of soft tissues.

Advantages

The advantages are as follows:

Strongest tested fixation.

May individually tension all bundles of graft.
 
Disadvantages


The disadvantages are as follows:

Pin may tilt in soft bone and lose fixation.

Steep learning curve of fiddle factor.

Special guides are required.



Buttons

Buttons are shown in Figure 6 and Figure 7.

Figure 6. The Endo-button periosteal cortical femoral fixation of hamstring grafts.



Figure 7 The periosteal button fixation of soft tissue grafts.

Advantages

The advantages are as follows:

The Endo-button with closed loop tape is strong, if expensive.
The plastic button is cheap, available and easy to do.
 
Disadvantages


The disadvantages are as follows:

Fixation site is distant with increase in laxity, with the bungee cord effect.

Increased in tunnel widening.

Plastic button has low pullout strength, dependent on the sutures.



Tibial Fixation

The tibial fixation remains a problem with soft tissue graft fixation. Patients generally do not tolerate metal devices in the subcutaneous area on the front of the tibia. The interference screw gets away from that problem, but has poor performance in cyclic load. The graft tends to slip out from under the screw as the knee is cycled. A backup fixation must be used it the interference screw is used. The Intrafix (Mitek) device uses the interference screw fixation principle, but increases both the ultimate load to failure and the cyclic load performance.

Considerations

The most important consideration in ACL reconstruction is that the tunnels are put in the correct position. After this, the fixation of the graft is the next most important factor in a satisfactory clinical outcome. The physician should become proficient at one of these techniques. For revi-sions, physicians may need to have available another type of fixation to deal with hardware and tunnel expansion.
 
Tendon-to-Bone Healing




Studies have shown that it takes at least 8 to 12 weeks for soft tissue to heal to bone, as compared to 6 weeks for bone-to-bone healing with the patellar tendon graft. Recent studies have shown that the compression of the tendon in the tunnel with a screw speeds the time of healing, similar to internal compression in bone healing.



Donor Site Morbidity

In 1982, Lipscomb found that after harvest of the semitendinosus only the strength of the hamstrings was 102% and after harvest of both the strength was 98%. Recently, it has been shown that the internal rotation strength is decreased after the harvest of the semitendinosus. The patellofemoral pain incidence has been reported by Aligetti to be 3 to 21% after semitendinosus reconstruction. There are rare reported cases of saphenous nerve injury.



Early Rehabilitation

Prospective randomized studies by Aligetti and Marder have shown that with early and aggressive rehabilitation, there was no difference between the semitendinosus and patellar tendon grafts in stability or final knee rating. This puts to rest the argument as to whether the hamstring graft can withstand early aggressive rehabilitation protocols.



Central Quadriceps Tendon

This graft has been largely ignored in North America over the past decade. An assistant can harvest the graft while the surgeon is doing the notchplasty. It is a large diameter graft, 10 ¥ 10mm (Fig. 5.12). The tendon graft is fixed with interference screws for the bone plug and sutures tied over buttons for the tendon end. A bioabsorbable interference screw may be used at the internal aperture of the tunnel to reduce the tendon motion in the tunnel. The quadriceps tendon graft should reduce the need for the allograft or synthetic in revision cases.
 

Figure. 8 The quadriceps tendon graft.

Allografts


Advantages

The allograft has no harvest site morbidity. With no harvest required, the time of the operative procedure is reduced.



Disadvantages



The main objection to the use of the allograft is the risk of disease transmission. Jackson has shown that it takes longer for the graft to incorporate and mature, meaning a longer time until the patient can return to sports. In addition, there is a limited availability of allograft materials. In the literature, Noyes has shown that in long-term follow-up, failure rates increase. In the 1997 survey of the ACL study group by Campell, none of the members used allografts for primary reconstructions.

Synthetic Grafts

The best scenario for the use of the LARS synthetic graft is when the graft can be buried in soft tissue, such as in extra-articular reconstruction. This allows for collagen ingrowth and ensures the long-term viability of the synthetic graft. It will be sure to fail early if it is laid into a joint bare, especially going around tunnel edges, and is unprotected by soft tissue.
Advantages


There is no harvest site morbidity with the use of the synthetic graft. The graft is strong from the time of initial implant. There is no risk of disease transmission.

Disadvantages

The main disadvantage is that all the long-term studies have shown high failure rate. There is the potential for reaction to the graft material with synovitis, as seen with the use of the Gore-Tex graft.With the Gore-Tex graft, there was also the increased risk of late hematogenous joint infection. The results that have been reported with the use of the Gore-Tex graft suggest that it should not be used for ACL reconstruction. Unacceptable failure rates have also been reported with the use of the Stryker Dacron ligament and the Leeds-Keio ligament. The ligament augmentation device was also found to be unnecessary.

 

Figure 9 The insertion of the BioScrew through the anteromedial portal

Figure 10. The insertion of the BioScrew into the femoral tunnel

WORK-RELATED BACK PAIN

Back pain, like tooth decay and the common cold, is an affliction that affects a substantial proportion, if not the entire population, at some point in their lives. Nobody is immune to this condition nor its potential disability which does not discriminate by gender, age, race or culture. It has become one of the leading causes of disability in our society and the cost of treatment has been increasing progressively each year, without any obvious effect on the frequency and severity of the condition. The search for a cure and the elimination of back pain does not appear to be a viable option at this point in our understanding of back pain. A reasonable goal, however, is to improve the ability of clinicians to determine the cause of back pain in a substantial proportion of patients, to identify conditions likely to lead to serious disability if not treated promptly, to reduce the symptoms of back pain, to increase functional capacity and to reduce the likelihood of recurrences.

EPIDEMIOLOGY

The prevalence of back pain in the adult population varies with age. There are a number of surveys in multiple countries that reveal a point-prevalence of 17–30%, a 1-month prevalence of 19–43% and a lifetime prevalence of 60–80%. The likelihood that an individual will recall on survey that they have experienced back pain in their lifetime reaches 80% by the age of 60 years, and there is some evidence that the remaining 20% have simply forgotten prior episodes of back pain or considered such episodes as a natural part of life and not worth reporting. At the age of 40 years, the prevalence is slightly higher in women, while, after the age of 50, it is slightly higher in men. The majority of these episodes of back pain are mild and short-lived and have very little impact on daily life. Recurrences are common and one survey found that up to 14% of the adult population had an episode of back pain each year that lasted 30 days or longer and at some point interfered with sleep, routine activities or work. Approximately 1% of the population is permanently disabled by back pain at any given point, with another 1–2% temporarily disabled from their normal occupation.

Children and adolescents are not immune from back pain. Surveys reveal that approximately 5% of all children have a history of back pain that interferes with activity, with 27% reporting back pain at some time.Normal spinal anatomy and physiology The spine is one of the most complex structures in the body. It is a structure that includes bones, muscles, ligaments, nerves and blood vessels as well as diarthrodial joints. In addition, the structures that make up the spine include the intervertebral discs, the nerve roots and dorsal root ganglia, the spinal cord and the dura mater with its spaces filled with cerebrospinal fluid. Each of these structures has unique responses to trauma, aging and activity.

WORK-RELATED BACK PAIN

Back injuries make up one-third of all work-related injuries or almost one million claims in the United States each year. Approximately 150 million workdays are lost each year, affecting 17% of all American workers. Half of the lost workdays are taken by 15% of this population, usually with prolonged periods of time loss, while the other 50% of lost work days are for periods of less than 1 week. The incidence rates for work-related back injuries vary, depending on the type of work performed. The factors that increase the likelihood of back injury are repetitive heavy lifting, prolonged bending and twisting, repetitive heavy pushing and pulling activities and long periods of vibration exposure. Work that requires minimal physically strenuous activity, such as the finance, insurance and service industries, has the lowest back injury rates, whereas work requiring repetitive and strenuous activity such as construction, mining and forestry has the highest injury rates.

PATHOLOGY AND BACK PAIN

There is a strong inclination on the part of clinicians and patients suffering from back pain, especially if it is associated with disability, to relate the symptoms of pain to pathological changes in spinal tissues. For this reason, there is a tendency to look for anatomical abnormalities to explain the presence of pain, by ordering X-rays, computerized tomography (CT) or magnetic resonance imaging (MRI) studies. It is tempting to point to changes in anatomical structure seen on these studies as the cause of the symptoms. Unfortunately, the assumption that the lesion seen on these studies is the cause of the pain is not always valid. Degenerative changes occur in virtually all patients as part of the normal aging process. At age 20, degenerative changes are noted on X-ray and MRI in less than 10% of the population. By age 40, such changes are seen in 50% of the asymptomatic population and, by age 60, this number reaches over 90%. Disc and joint pathology is noted in 100% of autopsies of persons over the age of 50. These changes can affect multiple levels of the spine and can be severe in the absence of symptoms. Pathology in the intervertebral disc can also exist in the absence of symptoms. Disc protrusion or herniation can be found in 30–50% of the population in the absence of symptoms. Even large and dramatic disc herniations and extrusions can be found in asymptomatic individuals. Changes in the intervertebral disc seen on discography, including fissures and radial tears, have recently been found to exist in patients without back pain. It is, therefore, not possible to interpret pathology seen on imaging studies as the origin of a person’s back pain without looking for other contributing factors or clinical findings.

PHYSIOLOGY OF BACK PAIN

There are a number of factors that have been implicated in the genesis of back pain and disability that can be used to determine whether a pathological process seen on imaging studies is associated with symptoms experienced by a patient. Certain of these factors are based on epidemiological studies, while others are based on clinical findings and physiological tests. Pain in any structure requires the release of inflammatory agents that stimulate pain receptors and generate a nociceptive response in the tissue. The spine is unique in that it has multiple structures that are innervated by pain fibers. Inflammation of the posterior joints of the spine, the intervertebral disc, the ligaments and muscles, meninges and nerve roots have all been associated with back pain. These tissues respond to injury by releasing a number of chemical agents that include bradykinin, prostaglandins and leukotrienes. These chemical agents activate nerve endings and generate nerve impulses that travel to the spinal cord. The nociceptive nerves, in turn, release neuropeptides, the most prominent of which is substance P. These neuropeptides act on blood vessels, causing extravasation, and stimulate mast cells to release histamine and dilate blood vessels. The mast cells also release leukotrienes and other inflammatory chemicals that attract polymorphonuclear leukocytes and monocytes.

These processes result in the classic findings of inflammation with tissue swelling, vascular congestion and further stimulation of painful nerve endings. The pain impulses generated from injured and inflamed spinal tissues are transmitted via nerve fibers that travel through the anterior (from nerves innervating the extremities) and posterior (from the dorsal musculature) primary divisions of the spinal nerves and through the posterior nerve roots and the dorsal root ganglia to the spinal cord, where they make connections with ascending fibers that transmit the pain sensation to the brain. The spinal cord and brain have developed a mechanism of modifying the pain impulses coming from spinal tissues. At the level of the spinal cord , the pain impulses converge on neurons that also receive input from other sensory receptors. This results in changes in the degree of pain sensation that is transmitted to the brain through a process commonly referred to as the ‘gate control’ system. The pain impulses are modified further through a complex process that occurs at multiple levels of the central nervous system. The brain releases chemical agents in response to pain known as endorphins. These function as natural analgesics. The brain can also block or enhance the pain response by means of descending serotonergic modulating pathways that impact with pain sensations both centrally and at the spinal cord level. The latter mechanism is felt to be responsible for the strong impact of psychosocial factors on the response to pain and the disability associated with back pain.

The pain centers in the spinal cord and brain can also change through a process known as plasticity which may explain the observation that many patients develop chronic pain that is more widespread than the pathological lesion and continues after the resolution of the peripheral inflammatory process.

APPROACHING THE PATIENT WITH BACK PAIN

The factors that determine the degree of back pain, and especially the amount of disability associated with the pain, are therefore the result of multiple factors. Structural pathology sets the stage and is the origin of the painful stimulus. The natural healing process, in most situations, results in the resolution of back pain within relatively short periods. Physical stress placed on the back through work and leisure activities may slow the healing process or irritate spinal pathology such as degenerative changes or disc protrusion. It is, however, the psychosocial situation of the patient that determines the level of discomfort and the response of a patient to the painful stimulus.

The patient’s psychological state, level of satisfaction with work and personal life as well as his/her social and spiritual life may impact upon the central modulation system in the brain and modify the response to pain. In this volume, a great deal of emphasis is placed on visualization of spinal lesions that can result in spinal pain. To rely on anatomical changes to determine the cause of back pain can, however, be very misleading to the clinician through the mechanisms described above. There are other examples in science that can be used as a model for looking at spinal pain. The Danish pioneer of quantum physics, Niels Bohr, claimed that science does not adequately explain the way the world is but rather only the way we, as observers, interact with this world. Early in the last century, it was discovered that light could be explained in terms of either waves or particles, depending on the type of experiment that was set up by the observer. Bohr postulated that it was the interaction between the scientist, as the observer, and the phenomenon being studied, in this case light, that was important. The same thing can be said forthe clinician approaching a patient with back pain.The conclusions reached by the clinician regarding the etiology of back pain in a specific case are often dependent on the interaction between the patient and the clinician and the training and experience brought to the decision-making process by both individuals.

There are other ways of looking at back pain. Chaos theory postulates that there is a delicate balance between disorder and order. The origin of the universe is generally explained by the ‘Big Bang’ theory which states that, in the beginning, there was total disorder which was followed by the gradual imposition of order through the creation of galaxies, stars and planets. This process is perceived as occurringthrough a delicate balance between the forces of gravity and the effects of the initial explosion. This process emphasizes that small changes at the beginning of a process or reaction can result in large changes over time. If one applies this analogy to the interaction between patients with back pain and their physicians, the outcome of treatment can be perceived as being impacted upon by a number of beneficial influences or ‘little nudges’ and harmful attitudes or ‘little ripples’ (Table 1). The patient’s symptoms can be positively impacted through such processes as listening, caring, laughter, explanation, encouragement, attention to detail and even prayer and negatively impacted by fear, anxiety, anger, uncertainty, boredom and haste. The manner in which a physician uses these nudges and helps the patient avoid the ripples can have a large effect on the impact of back pain on the patient’s life. The most accurate diagnosis possible is dependent on accurately observing and listening to the patient, the physical examination and the results of all testing in combination with the intuition that is gained from experience from treating multiple similar patients.

The fine balance between different factors impacting on back pain can be illustrated by a few simple examples.

Example 1

A 50-year-old woman presented to her doctor with symptoms and signs of a disc herniation confirmed by CT scan. She was the owner of a small cattle range and was worried about the condition of her animals. She underwent surgery to correct the disc herniation but her convalescence was prolonged for no apparent reason. After several months, the condition of her cattle herd improved and, at the same time, the patient’s symptoms improved. This raises the question as to the link between the patient’s symptoms, the disc herniation and the condition of her cattle.

Example 2

A 45-year-old gentleman in a position with a responsible insurance company presented to his doctor with symptoms and signs of severe L4–5 instability confirmed by stress X-rays. The patient underwent a posterolateral fusion. At 3 months, the fusion was solid but the patient’s symptoms did not improve. Further questioning revealed that he felt stressed and was unhappy in his work. At 6 months, he became symptom-free without further treatment. The only evident change in his status was the resolution of his difficulties at work.

Example 3

A 35-year-old gentleman with a wife and two small children was admitted to the hospital on an emergency basis with suspected cauda equina syndrome. A psychotherapist assigned to the case discovered that the patient found the presence of his mother-inlaw intolerable. Arrangements were made for the mother-in-law to live elsewhere and the patient made an uneventful recovery without the necessity of surgery.

Table 1 Beneficial influences (nudges) and harmful influences (ripples) which impact on the outcome of treament for back pain

Harmful influences                        Beneficial influences

Fear                                            Listening and caring
Anxiety                                       Laughter
Anger                                         Explanation
Uncertainty                                 Encouragement
Boredom                                    Attention
Haste                                         Prayer

 
THE BONY VERTEBRAE

Each of the bony elements of the back consist of a heavy kidney-shaped bony structure known as the vertebral body, a horseshoe-shaped vertebral arch made up of a lamina, pedicles and seven protruding processes. The pedicle attaches to the superior half of the vertebral body and extends backwards to the articular pillar. The articular pillar extends rostrally and caudally to form the superior and inferior facet joints. The transverse processes extend laterally from the posterior aspect of the articular pillar where it connects to a flat broad bony lamina. The laminae extend posteriorly from the left and right articular pillars and join to form the spinous process. Two adjacent vertebrae connect with each other by means of the facet joints on either side. This leaves a space between the bodies of the vertebrae which is filled with the intervertebral disc. The intervertebral foramen for the exiting nerve root is formed by the space between the adjacent pedicles, facet joints and the vertebral body and disc. The integrity of the nerve root canal is therefore dependent on the integrity of the facet joints, the articular pillars, the vertebral body endplates and the intervertebral disc. The bony vertebrae can be visualized on standard radiographs and on CT scan using X-radiation. The bones can also be visualized on MRI, although with not quite the same definition. The metabolism of the bony vertebra can be visualized by means of a technetium bone scan.

Figure 1.1 Superior view of an isolated lumbar vertebra

This view demonstrates the two posterior facets and the vertebral body endplate where the disc attaches. The facets and the disc make up the ‘three-joint complex’ of the spinal motion segment. The body of the vertebra is connected to the articular pillars by the pedicles. The superior and inferior articular facets extend from the articular pillars to connect with the corresponding facets of the vertebrae above and below, to make up the posterior facets. The lateral transverse processes and the posterior spinous process form the attachments for paraspinal ligaments and muscles.

THE INTERVERTEBRAL DISC
The intervertebral disc is made up of an outer annulus fibrosis and a central nucleus pulposus. It is attached to the vertebral bodies above and below the disc by the superior and inferior endplates. The nucleus pulposus  a gel-like substance made up of a meshwork of collagen fibrils suspended in a mucopolysaccharide base. It has a high water content in young individuals, which gradually diminishes with degenerative changes and wit the natural aging process. The annulus fibrosis is made up of a series of concentric fibrocartilaginous lamellae which run at an oblique angle of about 30º orientation to the plane of the disc. The fibers of adjacent lamella have similar arrangements, but run in opposite directions. The fibers of the outer annulus lamella attach to the vertebral body and mingle with the periosteal fibers. The fibrocartilaginous endplates are made up of hyaline cartilage and attach to the subchondral bone plate of the vertebral bodies. There are multiple small vascular perforations in the endplate,which allow nutrition to pass to the disc.
The intervertebral disc is not seen on standard Xray, but can be visualized by means of MRI scan and

CT scan. The integrity of the inner aspects of the disc is best visualized by injecting a radio-opaque agent into the disc. This material disperses within the nucleus and can be visualized radiologically as a discogram.

THE POSTERIOR FACETS

The facet joints connect the superior facet of a vertebra to the inferior facet of the adjacent vertebra on each side and are typical synovial joints. The articular surfaces are made of hyaline cartilage which is thicker in the center of the facet and thinner at the edges. A circumferential fibrous capsule, which is continuous with the ligamentum flavum ventrally, joins the two facet surfaces. Fibroadipose vascular tissue extends into the joint space from the capsule, particularly at the proximal and distal poles. This tissue has been referred to as a meniscoid which can become entrapped between the facets.
The posterior facets can be seen on X-ray but only to a limited extent. Degenerative changes and hypertrophy of the facets can be visualized to a greater extent on CT and MRI. Radio-opaque dye can also be injected into the joint and the distribution of the dye measured.

Figure 1.2 Lateral view of the L3 and L4 vertebrae

This projection demonstrates the manner in which the facets join. The space between the vertebral bodies is the location of the cartilaginous intervertebral disc. Courtesy Churchill-Livingstone (Saunders) Press

Figure 1.3 Transverse view of L2 showing normal intervertebral disc morphology

Figure 1.4 Longitudinal view of the lumbar spine showing normal disc size and morphology

Courtesy Churchill-Livingstone (Saunders) Press

Figure 1.5 Normal discogram

Ergonomia Physiological Effects of Back Belt Wearing During Asymmetric Lifting

Physiological Effects of Back Belt Wearing During Asymmetric Lifting

Thomas G. Bobick,* Jean-Louis Belard, Hongwei Hsiao, James T. Wassell
Division of Safety Research, National Institute for Occupational Safety and Health, Morgantown, WV 26505, USA



Abstract

This study investigated the effect of wearing a back belt on subjects' heart rate, oxygen consumption, systolic and diastolic blood pressure, and respiratory frequency during asymmetric repetitive lifting. Thirty subjects with materials-handling experience utilized three different belts (ten subjects per belt). Subjects completed six 30-minute lifting sessions — three while wearing a belt and three without. Data analyses were conducted on the second, third, and fourth lifting periods. A 9.4 kg box, without handles, was lifted three times/min, starting at 10 cm above the floor, ending at 79 cm, with a 60° twist to the right. Data analysis indicates that belt-wearing did not have a significant effect on the overall mean values for heart rate, systolic and diastolic blood pressure, and respiratory frequency. Belt-wearing had a significant effect on the overall mean oxygen consumption of the subjects.

Author Keywords: Manual lifting; Physiology; Back belt



1. Introduction

During the late 1980s and early 1990s, wearing back belts became quite popular in an attempt to reduce the number of low-back injuries occurring in the workplace. In 1994, the National Institute for Occupational Safety and Health (NIOSH) released a report (DHHS, 1994), which was a review of the published scientific literature, that stated that NIOSH "does not recommend the use of back belts to prevent injuries among uninjured workers, and does not consider back belts to be personal protective equipment." In an effort to evaluate the effectiveness of back belts, NIOSH planned and conducted a field study and two laboratory studies. The field study was a prospective epidemiological investigation of back belt usage in a national chain of retail establishments (Wassell et al., 2000). One laboratory study dealt with the biomechanical effects (Giorcelli et al., 2001) and the second one (the current study) dealt with the physiological effects of repetitive lifting while wearing a back belt.

The technical literature in the last 5-10 yr related to physiological (i.e., cardiovascular and respiratory) effects of lifting while wearing a back belt is quite limited. Two studies evaluated a traditional weightlifter's belt (Hunter et al., 1989) and a medical orthosis (Duplessis et al., 1998). Industrial workers do not use these types of back supports.

Five studies (Rafacz and McGill, 1996; Marley and Duggasani, 1996; Aleksiev et al., 1996; Soh et al., 1997; Rabinowitz et al., 1998) evaluated commercially available back belts typically used in industrial workplaces. Only one of the five studies ( Soh et al., 1997) evaluated more than one belt type. Three of the five studies required subjects to handle items that were of an industrial nature – a tote box ( Marley and Duggasani, 1996), a bucket ( Soh et al., 1997), and a beverage crate ( Rabinowitz et al., 1998). The other two studies used activities that were non-industrial. However, none of the studies investigated handling typical warehouse or grocery store boxes, which are smooth cartons, not equipped with handles or cut-out openings. Waters et al. (1993) indicate that the lifting capacity of workers is decreased when having to lift cartons without handles.

Basically, either a single belt was used to evaluate one or more physiological parameters (Marley and Duggasani, 1996; Rabinowitz et al., 1998) or multiple belts were used to evaluate a single physiological parameter, which was respiratory frequency ( Soh et al., 1997). The primary dependent variables that were investigated during the five studies included heart rate (HR), respiratory frequency (RF), oxygen consumption (2), systolic blood pressure (SBP), and diastolic blood pressure (DBP). Only one study (Marley and Duggasani, 1996) simultaneously evaluated the effect of belt-wearing on these primary physiological parameters.

Significant differences did not occur for HR, in any of the four studies that evaluated it, or for 2, which was investigated in a single study. A significant increase in RF occurred with only one of the three belts that were evaluated by Soh et al. (1997). Marley and Duggasani (1996) found a significant increase in both the SBP and DBP when wearing a back belt, but Rafacz and McGill (1996) found a significant increase in DBP only.

In summary, different results have been reported in the literature regarding the physiological effects of lifting when wearing a back belt. In addition, there is a void in the literature regarding the effect of wearing a back belt when lifting boxes not equipped with handles. Finally, only one previous study simultaneously evaluated all primary physiological parameters.

The objective of this research investigation was to determine whether wearing a back belt during the repetitive handling of a moderate-weight box, without handles, would have an effect on subjects' HR, RF, oxygen consumption, and blood pressure measurements.

2. Methods and materials

2.1. Subjects

A convenience sample of 30 subjects participated in this laboratory study. The test subjects were a mix of full-time employees of the Physical Plant of West Virginia University (n=20) and University students (n=10). All subjects had to have at least 3 months of prior materials-handling experience. The average age for the sample was 30.0 yr (sd=7.7), with an average height of 175.2 cm (sd 8.1), average weight of 81.1 kg (sd 14.7), and average body mass index (BMI) of 26.4 (sd 4.9). Five of the subjects were women and 25 were men. Table 1 provides a breakdown of age, height, weight, and BMI by gender. All subjects were free of low-back pain and other musculoskeletal complaints. An informed consent form was signed by each participant. Each subject was compensated for undergoing the physical and for participating in the test sessions.



Table 1. Anthropometric characteristics of test subjects

Subjects Number Age (yr)                 Height (cm)               Weight (kg)       Body mass index                                                   Mean (sd)              Mean (sd)                 Mean (sd)                 Mean (sd)

Men        25        30.1 (8.2)               177.5 (6.6)                84.6 (12.9)                 27.0 (4.9)

Women   5          29.6 (4.7)               163.6 (4.3)                63.3 (9.1)                   23.5 (3.7)

Total       30        30.0 (7.7)               175.2 (8.1)                81.1 (14.7)                 26.4 (4.9)



2.2. Experimental Design

The original design of this experiment was a within-subjects, repeated measures design, with six repeated trials per subject, so that in three randomly selected trials the subjects lifted while wearing a back belt, and in the other three, lifted without the belt. Based on a simulation study, which used data from Rafacz and McGill (1996), a sample size of 30 subjects was found to provide at least 80% power to detect a difference of 3.75 mmHg in the SBP. The order of belt- or no-belt-wearing was randomized across all subjects. Three different belts were used, with ten subjects assigned to each belt type. Once a belt was assigned, it was used throughout the test day. Because of uncontrolled variability with the lunch period, which is described in the Section 2.7, the analysis was conducted as a three-period crossover evaluation.

2.3. Test Method

After the subjects were recruited, they completed a health questionnaire that was managed by a medical doctor. If the initial health screening was acceptable, a clinical history was completed for each subject, and a physical exam, which included an EKG, was conducted by a medical doctor on staff with NIOSH. After passing the physical, each participant was scheduled for the 6-h test procedure. On the test day, the experimental protocol was thoroughly reviewed with each subject.

2.4. Test Procedure

The test protocol consisted of six 30-min lifting sessions — four were completed in the morning, the remaining two after a 75-min lunch break. A 15-min rest period was scheduled between each lifting session. A wooden box, without handles, measuring 30 cm wide × 25 cm deep (front to back) × 46 cm high, was grasped at opposing corners and lifted 3 times/min.

The lifts started at 10 cm above the floor (pallet height) and ended at 79 cm (table height), with a 60° twist to the right, relative to the subject's mid-sagittal plane at the beginning of the lift. In an attempt to mimic the real-world, the subjects were not told to use a specific lift type, such as a stoop lift (bent back) or a squat lift (bent knees). They used a free-style lifting technique that is typically used during normal work activities. Thus, slight variations in the subjects' lifting posture occurred from one lift to the next. Fig. 1 provides a view of the test set-up. Figs. 2 and 3 show the beginning and end of the lift, respectively.

Figure 1. Overall view of task lay-out, with lab assistant seated on right.

Figure 2. Beginning of lift.

Figure 3. End of lift.

Before lifting, the subjects could position their feet however it was comfortable for them. Their foot location was then marked (see Fig. 1) so they would start at the same location for all subsequent lifting sessions. They were instructed, however, that foot movement during lifting was not permitted. This was in accordance with the 1991 revised NIOSH lifting equation (Waters et al., 1994, p. 21).

The weight of the box, 9.4 kg, was determined with the 1991 revised NIOSH lifting equation (Waters et al., 1993), by using the physical parameters of the simulated industrial task that was just described. This equation provides guidance for weights to be lifted that will not exceed the acceptable lifting capacity of 99% of male workers and 75% of female workers ( Waters et al., 1993). The researchers felt that 9.4 kilograms, which corresponded exactly to the recommended weight limit of the revised equation (i.e., RWL=1.0), would be an appropriate weight to ensure that the subjects would not be at risk for an injury during the 6-h protocol.

Subjects were required only to lift the box. It was lowered to the starting location by a lab assistant, who is shown in Fig. 1.

2.5. Belt Types

The three belts utilized in this study were: (a) a woven mesh material that was stretchable and equipped with Velcro® closures on both sides for proper tightening, (b) a firm-woven belt, which was bendable but not stretchable, that had a 5-cm adjustable nylon strap (Velcro® on one side only) equipped with a plastic clip-lock buckle closure, and (c) a stretchable mesh material with two side panels of flexible rubber material, equipped with two 2.5-cm straps that fold back on themselves (after passing through ovals), using Velcro® for closing. This third belt type was worn lower on the pelvis than the other two.

2.6. Instrumentation

During the testing, subjects wore a heart rate monitor (Polar Vantage, Polar USA, Inc., Stamford, CT) against their skin. The monitor consisted of an adjustable latex chest belt that contained two electrodes. This monitor automatically sampled the subject's HR for every 12 s, and stored an average value every minute during the six 30-min lifting sessions. Subjects also wore a chest-mounted light-weight (0.8 kg) oxygen analyzer (CosMed K2, CosMed Corp., Rome, Italy) on the outside of their clothes. This portable device is an integrated telemetry system that measures RF and calculates minute ventilation and oxygen consumption. According to the documentation with the equipment, it has a reliability of ±10%. The face mask contains a mini-turbine respiratory flow meter and a capillary tube for sampling expired air, which is analyzed by a small polarographic electrode. Information is transmitted to a receiver unit that processes, archives, and displays data in real time.

Finally, blood pressure readings were collected at the beginning of each session, as well as at times 10, 20, and 30 min into the sessions. These measurements were taken manually during the 20-s pause between timed lifts by a medical doctor using a manual sphygmomanometer (Baumanometer, Baum, Inc., Copiague, NY).

2.7. Data Analysis

Before leaving for lunch, subjects were requested to avoid caffeinated beverages and refrain from smoking. However, the researchers noted a variety of personal activities that may have contributed to inconsistencies in the data collected in the afternoon, including smoking cigarettes, drinking caffeine beverages, eating spicy or filling foods, and having to hurry to return on time after lunch. The research team initially attempted to document what the subjects ate or drank at lunch, or whether they had to hurry back to the facilities after lunch. We realized that we could not adequately quantify these variations.

Since a number of subjects had already completed the testing, we decided that the afternoon sessions would be completed by all, but the data would not be included in the final analysis because of the extensive discrepancies among the subjects' lunch activities.

In addition, subjects did not have an opportunity to participate in any practice sessions prior to testing to become familiar with the lifting procedure and the physiological equipment to be worn. So, unbeknownst to the subjects, the first lifting period was used as a practice/warm-up session and was not included in the data to be analyzed.

Therefore, only data from the second, third, and fourth lifting periods were included in the analyses. This resulted in a more reliable data set for determining whether the back belt truly had an effect on the measured physiological parameters.

When conducting any type of aerobic activity (exercise, or work such as repetitive lifting), a person requires an initial period of time to ramp up to a steady-state condition of energy expenditure. Thus, only the final 20 min of the second, third, and fourth lifting periods were analyzed for this experiment. Heart rate, RF, and 2 data were collected each minute, and blood pressure readings were collected every 10 min during each of the three analysis periods.

A three-period crossover analysis was used (Jones and Kenward, 1989, Design 3.6.1, p. 175). Mean values and variances were calculated for each parameter in the three analysis periods, for each of the 30 subjects. A three-factor analysis of variance (subject, belt, and lifting period) was conducted using the mean values, as suggested by Crowder and Hand (1990), which were weighted by the inverse of the variance, from each of the three periods for all 30 subjects. All analyses included a check for a carryover effect, but none was found. The carryover parameter investigates whether a residual effect may have occurred (in a subsequent period) because the belt was worn in a prior period. The effect due to belt type was not analyzed because of insufficient statistical power.

Paired-comparison t-tests were conducted for HR, 2, and RF for each subject for the belt-on versus belt-off conditions. The Fisher exact test was used for both blood pressure values. Due to the random allocation of belt-wearing, one subject wore the belt during the three analysis periods and could not be used in the paired-comparison analyses. The data from this particular subject, however, was used in the estimate of the error variance in the three-period crossover analysis.

3. Results

3.1. Overall Group Data

Table 2 provides a summary of the mean values, F-statistics, and corresponding p-values from the ANOVA analyses for the five physiological parameters when a back belt was worn during the repetitive lifting of a moderate weight. Table 3 provides a summary of the overall statistically significant effects, both increased and decreased, that resulted from the paired comparisons for the five physiological parameters.



Table 2. Summary of mean values, F-statistics, and corresponding p-values from ANOVA analyses

Variable              Mean values   Mean values        F-statistics                  p-values

                           (Belt off)        (Belt on)            (with d.f. a =1.57)

HR (beats/min)    91.7              91.0                     0.5481                       0.4621

SBP (mmHg)       123.5           124.6                    3.1659                       0.0805

DBP (mmHg)       74.3             75.1                     2.8848                       0.0949

RF (breaths/min)  19.2              19.4                    0.7244                       0.3983

VO2 (l/min)          0.762           0.711                   4.6708                       0.0349

a d.f. = degrees of freedom.



Table 3. Summary of statistically significant effects (p>0.05), based on paired comparisons of no-belt-wearing versus belt-wearing, for five physiological parameters evaluated during repetitive lifting of a moderate-weight box.

Subjects         Heart rate         Respiratory frequency       BP systolic      BP diastolic       VO2

                     (beats/min)        (breaths/min)                    (mmHg)          (mmHg)             (l/min)
N = 29a            12                           3                                 0                      3                    9

                       8 n.s.                       18 n.s.                         29 n.s.             26 n.s.           13 n.s.

                         9                             8                                  0                      0                   7



a One subject wore the belt during the three analysis periods and could not be evaluated with paired-comparison analyses. See data analysis section for more detail.

= number of subjects showing a significant increase in the parameter when wearing the back belt.

= number of subjects showing a significant decrease in the parameter when wearing the back belt.

n.s. = number of subjects showing no significant difference in the parameter when wearing the back belt.

3.2. Heart Rate

Wearing the back belt did not result in a significant difference in the average HR of the subjects. The overall mean HR of the group for the no-belt condition and belt-wearing condition (p=0.462) was 91.7 and 91.0 beats/min, respectively. Paired-comparison t-tests were conducted for 29 of the 30 subjects. Table 3 shows that 12 subjects had significantly (p<0.05) larger mean HR values for belt-wearing (range=2.5 to 7.7 beats/min, mean=5.0 beats/min), and nine had significantly smaller mean HR values for belt-wearing (range=3.0 to 15.6 beats/min, mean=6.3 beats/min).

3.3. Respiratory Frequency

Wearing the back belt did not result in a significant difference in the average RF of the subjects. The overall mean group value for RF was 19.2 breaths/min for no-belt-wearing versus 19.4 breaths/min for belt-wearing (p=0.398). Paired-comparison t-tests for 29 subjects (Table 3) indicated that three had significantly (p<0.05) larger mean RF values for belt-wearing (range=0.8 to 2.5 breaths/min, mean=1.5 breaths/min), and eight had significantly smaller mean RF values for belt-wearing (range=1.0 to 3.1 breaths/min, mean=1.7 breaths/min).

3.4. Blood Pressure

Wearing the back belt did not result in a significant difference in the average values of both SBP and DBP. The overall mean group value for SBP for no-belt-wearing and belt-wearing (p=0.081) was 123.5 and 124.6 mmHg, respectively. The overall mean group value for DBP for no-belt-wearing and belt-wearing (p=0.095) was 74.3 and 75.1 mmHg, respectively.

The follow-up analysis used the Fisher exact test for SBP and DBP for each subject. None of the subjects had a statistically significant result for the systolic values (Table 3). Only three of the subjects had a significant difference (p<0.05) for the diastolic values when the belt was worn. These three had higher average DBP while wearing the belt (Table 3).

3.5. Oxygen Consumption

Wearing a back belt resulted in a statistically significant decrease in the average 2 of the subjects. The overall mean group value for 2 during lifting with no-belt-wearing was 0.762 l/min versus 0.711 l/min for belt-wearing (p=0.035). Paired-comparison t-tests conducted on the 29 subjects indicated that nine had significant increases (p<0.05) in mean 2 values for belt-wearing (versus no-belt-wearing) (range = 0.052-0.702 l/min, mean = 0.264 l/min, median = 0.099 l/min) and seven had significant decreases (p<0.05) in mean 2 values for belt-wearing (versus no-belt-wearing) (range= -0.033 to -0.370 l/min, mean= -0.125 l/min, median = -0.074 l/min).

4. Discussion

Results from the current study indicate that repetitively lifting a moderate-weight box, while wearing a back belt, did not have a significant effect on subjects' HR. This agrees with the studies conducted by Rafacz and McGill (1996), Marley and Duggasani (1996), Aleksiev et al. (1996), and Rabinowitz et al. (1998).

Similarly, the current study did not show a significant effect on subjects' RF when lifting while wearing a back belt. This agrees with the data from Marley and Duggasani (1996), but disagrees with the study by Soh et al. (1997) that found that one of the belts they evaluated had a significant increase in the RF. The researchers conjectured that the rigidity of that particular belt might have contributed to the increase in the respiratory frequency. Since the belt that was evaluated in the current study did not have a rigid construction, it seems quite logical that, in fact, there would not be an increase in RF.

Results from the current study indicate that wearing a back belt did not have a significant effect on the overall values for both SBP and DBP. These results disagree with the study by Rafacz and McGill (1996) and Marley and Duggasani (1996). This difference may be the result of the different types of tasks and work loads that were used in those studies. The present study used a moderate weight and involved a completely dynamic task. Rafacz and McGill (1996) used heavier weights and had isometric components as a part of their test procedure. Aleksiev et al. (1996) indicated that DBP was significantly (p<0.05) increased for isometric exertions versus dynamic lifting activities. The mechanical work load is greater in static work than in dynamic work because the heart has to work against the peripheral resistance caused by isometric exertions (Armstrong et al., 1980).

Marley and Duggasani (1996) found a significant increase in both SBP and DBP when wearing a back belt. That study had subjects lifting a tote box weighing 7 kg and 14 kg at 3, 6, and 9 lifts/min, with and without belt-wearing. The values for their 7 kg/3 lifts/min condition can be compared to the current study (9.4 kg/3 lifts/min). The data for the 7 kg/3 lifts/min resulted in a non-significant increase in SBP and DBP. This agrees with the current study.

Finally, the current study resulted in a significant decrease in 2 of the subjects when wearing the belt. These results disagree with the study by Marley and Duggasani (1996) which did not find a significant effect on subjects' 2 when wearing a belt while lifting.

The difference in 2 between belt-wearing (0.711 l/min) and no-belt-wearing (0.762 l/min) is 0.051 l/min (=51 ml/min=1.72 fluid oz/min). The resting metabolic rate for humans is about 3.5 ml O2 consumption/min/kg of body weight (deVries, 1986). Using the average body weight for the subjects in this study, which is shown in Table 1 to be 81.1 kg, results in a total of 284 ml O2 utilized/min when resting. Thus, the difference in 2 (i.e., 51 ml/min) is less than one-fifth of the 2 level needed by our "average" subject when resting.

Interestingly, the biomechanical study recently completed by NIOSH (Giorcelli et al., 2001) indicated that the subjects in that study bent their torso less and hips more, and they moved at a slower pace when wearing a back belt. Perhaps the slower lifting pace may have contributed to the slight reduction in the 2 consumed during the lifting while wearing the belt. A possible alternate explanation might be that the breathing pattern of the subjects was slightly altered when the belt was worn, thus causing the slight decrease in the 2 consumed during the lifting while wearing the belt. It should be mentioned that the subjects were instructed to wear the belt according to the manufacturer's instructions of "snug, but not too tight to be uncomfortable". None of the subjects complained during or after any of the six 30-minute lifting sessions that the belt felt uncomfortable.

A future study is suggested that will involve lifting a heavier weight. A heavier weight will increase the metabolic load, thus raising the HR and other physiological parameters, so any changes that may be caused by belt-wearing may be more obvious during the new test conditions. A future study may also want to evaluate an increase in the lifting frequency. Since a sizable portion of warehouse work involves handling weights heavier than 9.4 kg, or involves lifting boxes faster than 3 times/min, future studies that investigate one or both of these conditions would be valuable by simulating these more demanding conditions.

Finally, a future study that uses more subjects and multiple belt types may provide more insight regarding the effect of different belt types on cardiovascular and respiratory variables. Individual belt effects from the current study were not presented since there was not enough statistical power with a sample size of only 10 subjects per belt type.

5. Summary

This limited research study did not find an effect, either positively or negatively, in four of the five variables studied — HR, SBP, DBP, and RF— when subjects wore a back belt while repetitively lifting a moderate-weight box that did not have handles. This study indicated that a significant decrease occurred in the average 2 consumed by subjects when wearing the belt. The difference in 2 between wearing a belt and not wearing a belt is less than one-fifth of the value required by our "average" test subject when resting.

A future study that will use a heavier weight and/or an increase in the lifting frequency, involving more subjects and multiple belt types is suggested. Such a study may yield more detailed information on the effect of belt-wearing on subjects' cardiovascular and respiratory responses.



Acknowledgements

The authors gratefully acknowledge the assistance of Pam Graydon for her help with the scheduling, data collection, and test procedure activities, and to Ken Zara for his assistance with the testing procedure. A separate thank-you is extended to Cathy Inman, M.D. and Diana Freeland for their assistance in conducting the screening physicals. The authors also extend thanks to Nina Turner, Ph.D., Dan Habes, M.S., Chris Pan, Ph.D. (all of NIOSH), and Jerry Congleton, Ph.D. (Texas A&M University) for their review of the manuscript.



References

Aleksiev, A., Magnusson, M., Pope, M., Coblin, E., Luoto, S., 1996. Effect of back support on blood pressure during lifting. Poster Summary, International Society for the Study of the Lumbar Spine, June, Burlington, VT.

Armstrong, T.J., Chaffin, D.B., Faulkner, J.A., Herrin, G.D. and Smith, R.G., 1980. Static work elements and selected circulatory responses. Am. Ind. Hyg. Assoc. J. 41, pp. 254–260.

Crowder, M.J., Hand, D.J., 1990. Analysis of Repeated Measures, Chapman & Hall, London, NY.

deVries, H.A., 1986. Physiology of Exercise,4th Edition. Wm.C. Brown Publishers, Dubuque, IA.

Duplessis, D.H., Greenway, E.H., Keene, K.L., Lee, I.E., Clayton, R.L., Metzler, T., Underwood, F.B., 1998. Effect of semi-rigid lumbosacral orthosis use on oxygen consumption during repetitive stoop and squat lifting. Ergonomics 41, 790–797.

Giorcelli, R.J., Hughes, R.E., Wassell, J.T., Hsiao, H., 2001. The effect of wearing a back belt on spine kinematics during asymmetric lifting of large and small boxes. Spine 26, 1794–1798.

Hunter, G.R., McGuirk, J., Mitrano, N., Pearman, P., Thomas, B., Arrington, R., 1989. The effects of a weight training belt on blood pressure during exercise. J. App. Sports Sci. Res. 3(1), 13–18.

Jones, B., Kenward, M.G., 1989. Design and Analysis of Cross-over trials, Chapman & Hall, CRC, London, Boca Raton.

Marley, R.J., Duggasani, A.R., 1996. Effects of industrial back supports on physiological demand, lifting style, and perceived exertion. Int. J. Ind. Ergon. 17, 445–453.

Soh, T.N., Parker, P.L., Crumpton, L.L., Mealins, C., 1997. An investigation of respiration while wearing back belts. Appl. Ergon. 28, 189–192.

Rabinowitz, D., Bridger, R.S., Lambert, M.I., 1998. Lifting technique and abdominal belt usage: a biomechanical, physiological, and subjective investigation. Saf. Sci. 28, 155–164.

Rafacz, W., McGill, S.M., 1996. Wearing an abdominal belt increases diastolic blood pressure. J. Occup. Environ. Med. 38, 925–927.

U.S. Department of Health and Human Services (DHHS), 1994. Workplace use of back belts—review and recommendations. DHHS (NIOSH) Pub. No. 94–122, 28 p.

Wassell, J.T., Gardner, L.I., Landsittel, D.P., Johnson, J.J., Johnson, J.M., 2000. A prospective study of back belts for prevention of back pain and injury. JAMA 284, 2727–2732.

Waters, T.R., Putz-Anderson, V., Garg, A., Fine, L.J., 1993. Revised NIOSH equation for the design and evaluation of manual lifting tasks. Ergonomics 36, 749–776.

Waters, T.R., Putz-Anderson, V., Garg, A., 1994. Applications manual for the revised NIOSH lifting equation. DHHS (NIOSH) Pub. No. 94-110, 153pp. (Available from Department of Commerce, NTIS, Springfield, VA 22161, (703) 487-4650, ask for PB94-176930).



* Corresponding author. Tel.: +1-304-285-5986; fax: +1-304-285-6047; email: txb4@cdc.gov



Applied Ergonomics
Volume 32, Issue 6
December 2001
Pages 541-547



Back