Search Menu


Subgrouping Patients with Low Back Pain: What Group is Most Likely to Benefit from Pilates?

By Tom Lavosky, DPT, Cert. MDT

As seen in the Spring 2009 Balanced Body Pilates COREterly

Physical therapists utilize a wide range of interventions in the management of patients with low back pain (LBP). Approximately 50% of all patients presenting to outpatient physical therapy clinics seek treatment for LBP. Clinical research has not shown that one treatment is superior to another1 where “broad inclusion criteria”, meaning anyone with low back pain (LBP), is used.

Recent research, however, does show that if patients with LBP are placed in subgroups based on a cluster of signs and symptoms, one treatment does demonstrate greater efficacy over others.2

From this research, a classification system consisting of four groups has been developed. Choosing an intervention based on certain criteria is what is known as a clinical prediction rule (CPR). Inclusion criteria for each of the groups discussed below is not an exhaustive list, although those listed are considered to be the most important.


The first group is patients who are most likely to benefit from manipulation. The criteria for this group include the following: pain does not extend from the back to below the knee, symptoms are of recent onset (< 16 days), and decreased spinal mobility with spring testing. During a manipulation, an audible pop often occurs, which is known as a cavitation. These forces produce a barrage of input into the nervous system often resulting in pain inhibition.

Even though it is not shown in the literature, it is my belief that acupuncture has similar classification criteria as manipulation, i.e. short duration of symptoms and the absence of symptoms radiating below the knee. This is based on both a personal experience, for I received two acupuncture treatments after an acute episode of LBP with significant improvement, and anecdotal accounts by patients who were successfully treated with this intervention.


The second group, known as the stabilization classification, consists of patients who are likely to benefit from strengthening of deep abdominal and spinal musculature. Examples of deep trunk muscles include the transverse abdominis (TrA) and multifidus. The TrA encompasses the entire torso, essentially forming a corset. In Pilates’ parlance these deep structures are referred to as the powerhouse. In asymptomatic individuals, the TrA functions tonically and is noted to be the first trunk muscle activated during an anticipated perturbation of the spine.3 In patients with LBP, however, the anticipatory activation of the TrA becomes either delayed or absent. The rationale for incorporating Pilates in the rehabilitation program for patients in this group is to retrain the TrA to function tonically, rather than phasically. Inclusion criteria for this group include the following: <40 years old, chronic LBP or increasing episode frequency, excessive flexibility, positive prone instability test, and an instability catch or aberrant movement with bending forward or backward.

Specific Exercise

The third group consists of patients who are likely to benefit from a specific exercise. Classification criteria for this group include the following: symptoms extending below the buttocks, centralization with repeated motion testing, i.e. abolishment of symptoms from an area more distal to the buttocks, to a location more proximal or closer to the midline of the lower back,4 preference for either sitting with an arch in the lower back or sitting slouched, and peripheralization of symptoms with repeated motion testing when the direction is opposite to centralization, i.e. symptoms move from a location closer to the middle of the lower back to a more distal location, such as the thigh or calf.4

(continued above)

An example of centralization and pheriperalization is as follows: a patient has lumbar pain that radiates down his left buttock through the back of his thigh and into his calf. After repeatedly bending backward, he now reports that the pain in his calf and thigh are no longer present and has moved up to his buttocks. Conversely, when he repeatedly bends forward or sits slouched for an extended period of time, the pain extents from his buttocks through the back of the thigh and into the calf. This example illustrates a plausible scenario in a patient with a posterior disc protrusion without a relevant lateral component. This classification is based on the McKenzie method in which centralization is the desired outcome.


Finally, the fourth group is comprised of patients who may benefit from a trial of traction. Classification criteria for this group include the following: signs and symptoms of nerve root compression, i.e. symptoms extending below the knee, diminished Achilles or knee reflex, and decreased strength of a particular muscle group in the leg, and no centralization of symptoms with motion testing. This group is the least likely to benefit from conservative intervention and often requires a surgical decompression procedure if symptoms continue to worsen.

Ostensibly, not all patients with LBP can be placed in one of these four groups. LBP can be the result of a serious pathology such as an infection, inflammation, or visceral disease. If there is suspicion of symptoms resulting from a non-mechanical pathology, patients should be referred to a physician for medical work-up.

In conclusion, using this classification system can greatly improve the outcomes of patients with LBP. The greater the number of criterion that are present for a particular group, the more likely the intervention will have a successful outcome.


  • Delitto A. Research in low back pain: time to spot seeking the elusive “magic bullet.” Phys Ther. 2005;85:206-208.
  • Brennan GP et al. Identifying subgroups of patients with acute/sub-acute “nonspecific low back pain: results of a randomized clinical trail. Spine. 2006;31:623-631.
  • Hodges P.W. Is there a role for transverses abdominis in lumbo-pelvic stability? Manual Therapy. 1999;4(2):74-86.
  • Aina A, May S, Clare H. The centralization phenomenon of spinal symptoms -- a systematic review. Man Ther. 2004;9:134-143.