Research Review By Dr. Michael Haneline©

Date Posted:

Sept. 2009

Study Title:

The Flip Test


Summers B et al.

Author's Affiliations:

Department of Orthopaedics, The Princess Royal Hospital, Telford, Shropshire, UK.

Publication Information:

Spine 2009; 34(15):1585–89.

Background Information:

A commonly used test to determine the presence of a sciatic nerve lesion is the flip test. The test was first described in 1958 in a paper that was essentially an opinion piece that included no references.

The performance of the flip test was originally described as having the patient sit erect on an examining table with their legs dangling. The examiner then places one hand on the distal thigh to stabilize the patient, while the other hand grasps the heel of their affected leg and gradually extends the knee.

If the patient has true sciatic tension, there will be no resistance or leg pain until 45° of leg movement has been passed. Extending the leg past that point results in the patient falling backward (or “flipping”), often needing to brace themselves against the table with their hands in order to keep from falling backwards.

Various depictions of the test have appeared in text books and other publications, which typically include a description of a sudden backwards movement of the trunk that occurs in patients with true nerve root tension. However, as pointed out by the authors, the flip test had never been validated before this study.

Because of this and the fact that it is commonly used in clinical practice (and advocated by Waddell in his famous publications), this study’s purpose was to assess the test’s construct validity by comparing it with the supine straight leg raise test (SLR) in patients with MRI-confirmed nerve root compression.

Pertinent Results:

NOTE: the complete study methods are descrived below, but the following excerpt will help with the understanding of this section: Group 1 patients reported no pain and exhibited no bodily or verbal reaction to the Flip test. Group 2 patients reported pain in the leg and/or back on full knee extension. Group 3 patients reported pain in the leg and/or back and would not allow the knee to fully extend.

Some patients from Groups 2 and 3 reacted to knee extension by leaning the trunk back 10° to 20° and placing their hands beside or behind their buttocks to brace themselves. Some patients even lifted their buttocks off the table using their hands or by rolling back on their pelvis. Most of the time (except in 1 patient) this movement was gradual with no sudden or dramatic falling back. Three patients leaned backward but did not brace with their hands.

The patients’ average degrees of hip flexion during the supine SLR were compared to their flip responses using a one-way analysis of variance (ANOVA) and t-tests:
  • The mean degrees of supine SLR between groups 1 and 3 were significantly different, via ANOVA statistical analysis, as one would expect.
  • There were significant differences between the mean degrees of supine SLR among groups 1 and 2, and 2 and 3 using unpaired t-tests.
  • There was no significant difference in the percentage of patients exhibiting the lean back/hands brace reaction between groups 2 and 3 using the ?2 test.
  • There was no significant difference in the supine SLRs of patients in groups 2 and 3 who exhibited the lean back/hands brace reaction compared with those in groups 2 and 3 without the reaction.
A kappa analysis was performed in order to determine the level of agreement between the degrees of supine SLR and positive flip tests. Calculations were made using different supine SLR values in order to find the cut-off point that resulted in the highest kappa value, which was 48°/49°. This produced a kappa of 0.771 (95% confidence interval: 0.611 to 0.932), which points to moderate to substantial agreement between the 2 tests.

Another cut-off point for the degrees of supine SLR was identified as being of most value in the clinical setting. This value was set at 45° and all patients who had a supine SLR < 45° reacted painfully on the flip test, whereas the test was inconsistent and therefore not as valid in patients who had a supine SLR ? 45°.

Clinical Application & Conclusions:

The authors concluded that the description of the flip test that was given by Michele was more accurate than what others have provided. However, they indicated that the test was not capable of confirming the presence of nerve root tension, as originally espoused by Michele.

In the current study, only 1/3 of the patients exhibited a classic “flip test” where the patient responded by suddenly leaning back and bracing themselves on the exam table with their hands. Another 1/3 of the patients did not react to the test at all. This was evident in spite of the fact that all patients in the study had examination findings which indicated nerve root compression and was confirmed by MRI.

They suggested that the flip test would be useful in patients with supine SLRs that were < 45° to verify the presence of nerve root tension. Its use would be very limited as an indicator of nerve root tension in patients with supine SLRs ? 45°.

The authors also pointed out that the flip test had been erroneously used in the past by many doctors who wrongly classified their patients as not having sciatica when they really did. Some patients have even been labeled malingerers or fakers because the result of their flip test was at odds with the supine SLR (this test was included in the series of Waddell’s tests that we should all be familiar with).

The bottom-line of this paper is that the flip test would be useful in clinical practice to confirm the presence of sciatica in selected patients (i.e., those with supine SLRs that were < 45°), but it should never be used to rule out sciatica… especially to go so far as to label a patient as a malingerer.

Study Methods:

Patients were included in this study if they had:
  • signs and symptoms of sciatica (i.e., posterior leg pain radiating below the knee),
  • a significant disc protrusion evident on MRI that correlated with the signs and symptoms of sciatica, and
  • a restricted SLR that aggravated their leg pain
Patients were excluded if they:
  • were below the age of 18 years or above the age of 65 years, or
  • had undergone previous back surgery
Patients that met the above criteria were consecutively recruited from the clinical practice of the senior author. A total of 67 patients were recruited, including 35 men and 32 women, average 42 years of age. On average, patients had their symptoms for 6 months prior to the study.

Significant disc protrusions were mostly present at the L5–S1 level (41 patients), but 25 were found at L4–L5 and 1 at L3–L4. The protrusions were located on the right side in 29 patients and the left side in 38 patients.

Patients were initially administered a supine SLR test, which was measured using a long arm goniometer. Immediately afterward, they were placed in a seated position with their legs hanging over the edge of the couch and their knee was gradually extended, noting the patients’ reaction (i.e., they underwent the flip test).

In addition to noting whether patients exhibited a leaning back/hands braced reaction, patients were grouped according to 1 of 3 possible responses associated with knee extension: Group 1 patients reported no pain and exhibited no bodily or verbal reaction. Group 2 patients reported pain in the leg and/or back on full knee extension. Group 3 patients reported pain in the leg and/or back and would not allow the knee to fully extend.

Study Strengths / Weaknesses:

The authors’ concept of assessing the validity of a commonly used clinical test that appeared to function differently in clinical practice than what has been published is noble.

There were a number of weaknesses in this study, as follows:
  • There was no randomization of subjects; rather they were consecutively presenting patients.
  • All of the subjects had the condition that the test was designed to detect (i.e., sciatica), which is a poor way to conduct a study that assesses the validity of a test. It is much better to include people with and without the condition, as well as those with differing degrees of the condition’s severity.
  • Only one examiner was involved.
  • The examiner knew the results of the initial test (i.e., the supine SLR), which can result in “expectation bias.” (1)
  • The examiner knew the history of each patient (i.e., that they had sciatica).
Because of these weaknesses, the study’s findings cannot wholly be accepted as true and we must wait for further studies to confirm it. However, what the authors pointed out about the inconsistencies of the flip test and its unproven validity are worth considering.

Additional References:

  1. Greenhalgh, T. How to read a paper: Papers that report diagnostic or screening tests. BMJ 1997; 315(7107): 540-543.