British Journal of Rheumatology 1990;29:268-273
INFRARED THERMOGRAPHIC IMAGING, MAGNETIC RESONANCE IMAGING, CT SCAN AND MYELOGRAPHY IN LOW BACK PAIN BY D. THOMAS*, D. CULLUM*, GEORGIA SIAHAMIS* AND SUZANNE LANGLOISt 'Adelaide Medical Thermography Centre, 55 Jerningham Street, North Adelaide, South Australia; Adelaide Hospital, North Terrace, Adelaide, South Australia
fRoyal
KEY WORDS:
Imaging, Spine, Skin temperature, Lumbar nerve root, Comparisons, Thermography.
ing, exercise, physiotherapy or acupuncture on the morning of the test. Patients were advised to shower and remove cream or cosmetics over areas to be thermographed. Clothes were removed and equilibration occurred in a room thermostatically cooled to 21°C for 20 min prior to IRT assessment. Four areas were evaluated: lumbar spine and buttocks, anterior and posterior aspects of legs, and plantar area of feet. Thermographic temperature assessments were performed on 14 areas (seven each side) by computer analysis using boxes of standardized size for each area. The areas were gluteal, anterior and posterior areas of thighs, anterior and posterior areas of forelegs, dorsal and plantar surfaces of feet. The side-to-side temperature difference was computed for six specific regions as follows: (1) anterior leg (summation of thigh, foreleg and dorsal foot areas); (2) posterior leg (summation of gluteal, posterior thigh, calf and plantar foot areas); (3) gluteal; (4) plantar foot; (5) anterior foreleg; (6) anterior thigh. It was assumed that region 1 (anterior leg) represented more L5 root than region 2 (posterior leg), which contained more SI root. Region 3 (gluteal) covered S2 and S3 roots, region 4 (plantar foot) L5 and SI roots, region 5 (anterior foreleg) L4 and L5 roots and region 6 (anterior thigh) L3 and L4 roots. Thermographic analysis of the lumbar area was recorded visually; abnormal if there was a unilateral heat flare, a general increase centrally over the lower lumbar area or with bilateral heat flares. The leg thermograms were also examined visually and recorded abnormal with significant side-to-side thermal asymmetry. For computer based side-to-side temperature difference (Af°C), if the symptomatic side was colder a negative prefix was given before r°C and a positive prefix if the temperature on the symptomatic side was hotter. MRI assessments were performed using a Siemens Magnetom, using a super conductive magnet of 1 tesla strength cooled with liquid helium and nitrogen. T2 weighted sagittal scans and Tl weighted axial scan views were taken of the lumbar spine. All MRI scans were reported by SL. The significance of differences between group means was assessed by paired and nonpaired Student's f-tests.
COMPUTERIZED infrared
thermography (IRT) and magnetic resonance imaging (MRI) are useful in the assessment of low back pain. Thermography measures local skin temperature and therefore provides physiological data mediated through sympathetic nerve fibres. Characteristic skin surface temperature changes occur with lumbar nerve root irritation [1], non-radicular referred leg pain [2], and with lumbar disc pathology [3]. The diagnostic information shows significant correlation with CT scan, myelography [4] and MRI [5]. Magnetic resonance imaging provides fine anatomical details of the lumbar disc, nerve root and associated structures and can reveal changes which may have clinical relevance. In the present study, both visual and computerized IRT of patients with chronic lumbar pain with and without leg discomfort were compared prospectively with MRI and previously performed CT scans, myelography and discography. Many of the reported comparative studies between IRT, MRI and conventional radiological tests have relied largely on visual assessments of side-to-side leg temperature differences rather than more objective computer derived temperatures. PATIENTS AND METHODS Sixty-five patients with chronic back pain with and without referred leg pain (radicular or non-radicular) were assessed prospectively with MRI and IRT. Sixtyone of these were also examined by lumbar CT scan, 41 by myelography and 12 by discography. Thirty-seven were male and 28 were female, with a mean age of 41 (range 23-71 years). The mean duration of back pain at initial assessment was 2 years (range 2 months to 14 years). Thermographic assessments were performed with an Agema 782 infrared system using CATS (computer assisted thermographic software) and an IBM PXT for digitization of signals, storage and analysis of data. All patients followed a strict preparation protocol prior to IRT involving fasting for 3 h with no analgesics, smokSubmitted August 1989; revised version accepted 1 January 1990. Correspondence to Dr D. Thomas. 268
Downloaded from http://rheumatology.oxfordjournals.org/ at University of Iowa Libraries/Serials Acquisitions on May 29, 2015
SUMMARY Sixty-five cases of chronic low back pain were studied. Infrared thermography (IRT) was abnormal in 92%, magnetic resonance imaging (MRI) in 89%, computerized tomography (CT) in 87% and myelography in 80%. IRT correlated with MRI in 94% of cases, and with CTin87% of cases. Of 22 MRI positive disc and root cases, 21 (95%) had significant leg abnormalities on IRT. All 19 cases with radicular involvement on CT and all 18 with radicular involvement on myelography demonstrated significant leg changes on IRT.
THOMAS ETAL.: IRT, MRI, CT AND MYELOGRAPHY IN LOW BACK PAIN
l(a).
FIG.
l(b).
FIG.
3.
FIG. 1(C).
FIG.
2.
FIG. 1.—Thermography images, (a) Lumbar and gluteal showing a left lumbar flare, (b) Anterior leg (left cooler), (c) Plantar feet (left cooler). FIG. 2.—Magnetic resonance image of patient in Fig. 1. Transverse view at L4-5 level showing a left prolapse (arrowed) with extensive fibrosis extending from the theca laterally around the left L5 nerve root. FIG. 3.—Myelographic image of same patient. Left lateral obligue view. Largefillingdefect (arrowed) and non-filling of left L5 root.
Downloaded from http://rheumatology.oxfordjournals.org/ at University of Iowa Libraries/Serials Acquisitions on May 29, 2015
FIG.
269
270
BRITISH JOURNAL OF RHEUMATOLOGY VOL. XXIX NO. 4
OVERALL CHANGES FOR
TABLE I MRI, IRT, CT, MYELOGRAPHY
AND DlSCOGRAPHY
MRI (%)
IRT (%)
CT (%)
Myelography (%)
Discography (%)
Normal Abnormal Total
7 (11) 58 (89) 65 (100)
5 (8) 60 (92) 65 (100)
8 (13) 53 (87) 61 (100)
8 (20) 33 (80) 41(100)
1 (8) 11 (92) 12(100)
Abnormal* Disc only Disc & root Root only Total
36 (62) 22 (38)
12 (20) 32 (53) 16 (27) 60(100)
34 (64) 19 (36)
15 (45) 17 (52) 1 (3) -(100)
11(100) — — — — 11 (100)
58(100)
53(100)
* Lumbar and leg instead of disc and root for IRT group.
Four with bilateral referred leg pain and two with bilateral nerve involvement were excluded from subgroup analysis. Also excluded was a case with proven TB of the right sacroiliac joint and another with combined left sciatica and knee pain after a fracture. RESULTS Figures 1-3 illustrate a case of discogenic back pain and left sciatica after a discectomy at the L4-5 level, using imaging modalities employed in this study. Sixty-five cases of back pain were assessed by MRI and IRT and the results are summarized in Table I. The nerve root changes comprised loss of epidural fat, definite nerve root displacement or compression by postoperative fibrous tissue. Comparisons between MRI and IRT results are shown in Table II. An overall correlation between MRI and IRT was present in 89%. Lumbar CT scans were abnormal in 87% of cases (Table I). Comparisons between CT and IRT changes are shown in Table III. All cases with radicular involvement on CT show definite thermal asymmetry in the legs. There was an overall correlation between CT and IRT in 87%. Lumbar myelograms were abnormal in 80% of cases (Table I). Comparisons between myelography and IRT results are shown in Table IV. In all of the 18 cases where myelography demonstrated nerve root involvement, IRT demonstrated significant thermal asymmetry in the legs. Twelve also demonstrated lumbar changes on IRT. Twelve discograms were performed and 11 were abnormal (92%), showing leakage of dye with typical back pain (Table I). All were abnormal on IRT.
Objective analysis of thermal data For the purpose of more detailed computerized thermographic analysis, three well defined back pain groups were formed. Group 1, consisted of 21 cases with unilateral radiculopathy and discogenic back pain as defined by MRI and or CT and myelography; group 2, 12 with central discogenic back pain without unilateral or bilateral radiculopathy or symptomatic leg pain; group 3, 18 with lumbar disc degeneration and non-radicular referred leg pain. i.e. there was no nerve root involvement on MRI, CT and myelography. Thermal asymmetry The mean side-to-side temperature differences and standard deviation for the six specific thermographic regions for groups 1 and 2 is given in Table V and for group 3 in Table VI column 2. The mean temperature differences shown demonstrate that the symptomatic side of the body was cooler than the asymptomatic side. Side-to-side temperature difference in group 1 (unilateral radiculopathy cases) was maximal in the posterior leg region (—1.33°C) followed by the anterior leg region (-0.92°C) then the plantar foot region (—0.75°C). In group 2 (discogenic non-radicular nonreferred leg pain cases) the side-to-side temperature differences for all specific regions were negligible. Comparison between groups 1 and 2 demonstrated statistically significant differences for anterior leg, posterior leg, plantar foot, and anterior foreleg regions (Table V) and no significant differences for gluteal and anterior thigh areas. In group 3 (discogenic non-radicular referred leg pain cases) the side-to-side temperature differences were smaller than in group 2 but showed similar regional temperature distributions.
TABLE II
TABLE III
M R I COMPARED WITH I R T FOR DiSC AND NERVE ROOT CHANGES
C T COMPARED WITH IRT FOR DiSC AND NERVE ROOT CHANGES
Infrared thermographic abnormalities MRI changes
N
Disc only Disc & root Total
36 22 58
Lumbar only
Leg only
12 (33%) 8 (22%) 1 (5%) 6 (27%)
% relate to specific MRI subgroups.
Lumbar &leg
Infrared thermographic abnormalities
Normal
CT changes
N
15 (42%) 1 (3%) 15 (68%) 0
Disc only Disc & root Total
34 19 53
Lumbar only
Leg only
Lumbar &leg
Normal
9 (26%) 0
9 (26%) 6 (32%)
15 (44%) 13 (68%)
1 (3%) 0
% relate to specific CT subgroups.
Downloaded from http://rheumatology.oxfordjournals.org/ at University of Iowa Libraries/Serials Acquisitions on May 29, 2015
Overall changes
THOMAS ETAL.: IRT, MRI, CT AND MYELOGRAPHY IN LOW BACK PAIN TABLE IV MYELOGRAPHY COMPARED WITH IRT FOR DISC AND NERVE ROOT CHANGES
Infrared thermographic abnormalities N
Disc only Disc & root Root only Total
15 17 1 33
Lumbar only
Leg only
7(46%) 0 0
4(27%) 5(29%) 1 0
Lumbar &leg
Normal
4(27%) 12(71%)
— 0 0
% relate to specific myelography subgroups.
Comparisons between groups 1 and 3 demonstrated statistically significant differences for side-to-side temperatures in the posterior leg region only (P value 0.05
NS, not significant; Sign, significant; Af°C, mean symptomatic to asymptomatic side temperature difference.
TABLE VI GROUP 1 LUMBAR DISC DEGENERATION WITH UNILATERAL RADICULOPATHY (RADIC) (21 CASES) COMPARED WITH GROUP 3 LUMBAR DISC DEGENERATION AND NON-RADICULAR REFERRED LEG PAIN (REF. PAIN) 18 CASES MEAN SIDE-TO-SIDE TEMPERATURE DIFFERENCE
Radic
Ref. pain
Thermographic region
A/°C
(lSD)
At°C
(lSD)
(Test
Significance
Anterior leg Posterior leg Gluteal Plantar foot Anterior foreleg Anterior thigh
-0.92 -1.33 -0.18 -0.75 -0.47 -0.15
(0.92) (0.82) (0.33) (0.52) (0.49) (0.34)
-0.52 -0.56 -0.07 -0.43 -0.21 -0.06
(0.6) (1.19) (0.29) (0.74) (0.36) (0.32)
P>0.l P0.l
NS Sign NS NS NS NS
NS, not significant; Sign, significant; Ar°C, mean symptomatic to asymptomatic side temperature difference.
Downloaded from http://rheumatology.oxfordjournals.org/ at University of Iowa Libraries/Serials Acquisitions on May 29, 2015
Myelography changes
271
272
BRITISH JOURNAL OF RHEUMATOLOGY VOL. XXIX NO. 4 TABLE VII
GROUP 3 LUMBAR DISC DEGENERATION WITH NON-RADICULAR REFERRED LEG PAIN (REF. PAIN) 18 CASES, COMPARED WITH GROUP 2 LUMBAR DISC DEGENERATION WITHOUT RADICULOPATHY OR REFERRED LEG PAIN (NON-RADIC) 12 CASES. MEAN SIDE-TO-SIDE TEMPERATURE DIFFERENCES
Ref. pain
Non-radic
Ar°C
(lSD)
Ar°C
(lSD)
Anterior leg Posterior leg Gluteal Plantar foot Anterior foreleg Anterior thigh
-0.52 -0.56 -0.07 -0.43 -0.21 -0.06
(0.6) (1.19) (0.29) (0.74) (0.36) (0.32)
+0.10 -0.02 -0.10 +0.04 +0.05 +0.04
(0.22) (0.27) (0.23) (0.26) (0.25) (0.19)
f-Test
Significance
P
INFRARED THERMOGRAPHIC IMAGING, MAGNETIC RESONANCE IMAGING, CT SCAN AND MYELOGRAPHY IN LOW BACK PAIN BY D. THOMAS*, D. CULLUM*, GEORGIA SIAHAMIS* AND SUZANNE LANGLOISt 'Adelaide Medical Thermography Centre, 55 Jerningham Street, North Adelaide, South Australia; Adelaide Hospital, North Terrace, Adelaide, South Australia
fRoyal
KEY WORDS:
Imaging, Spine, Skin temperature, Lumbar nerve root, Comparisons, Thermography.
ing, exercise, physiotherapy or acupuncture on the morning of the test. Patients were advised to shower and remove cream or cosmetics over areas to be thermographed. Clothes were removed and equilibration occurred in a room thermostatically cooled to 21°C for 20 min prior to IRT assessment. Four areas were evaluated: lumbar spine and buttocks, anterior and posterior aspects of legs, and plantar area of feet. Thermographic temperature assessments were performed on 14 areas (seven each side) by computer analysis using boxes of standardized size for each area. The areas were gluteal, anterior and posterior areas of thighs, anterior and posterior areas of forelegs, dorsal and plantar surfaces of feet. The side-to-side temperature difference was computed for six specific regions as follows: (1) anterior leg (summation of thigh, foreleg and dorsal foot areas); (2) posterior leg (summation of gluteal, posterior thigh, calf and plantar foot areas); (3) gluteal; (4) plantar foot; (5) anterior foreleg; (6) anterior thigh. It was assumed that region 1 (anterior leg) represented more L5 root than region 2 (posterior leg), which contained more SI root. Region 3 (gluteal) covered S2 and S3 roots, region 4 (plantar foot) L5 and SI roots, region 5 (anterior foreleg) L4 and L5 roots and region 6 (anterior thigh) L3 and L4 roots. Thermographic analysis of the lumbar area was recorded visually; abnormal if there was a unilateral heat flare, a general increase centrally over the lower lumbar area or with bilateral heat flares. The leg thermograms were also examined visually and recorded abnormal with significant side-to-side thermal asymmetry. For computer based side-to-side temperature difference (Af°C), if the symptomatic side was colder a negative prefix was given before r°C and a positive prefix if the temperature on the symptomatic side was hotter. MRI assessments were performed using a Siemens Magnetom, using a super conductive magnet of 1 tesla strength cooled with liquid helium and nitrogen. T2 weighted sagittal scans and Tl weighted axial scan views were taken of the lumbar spine. All MRI scans were reported by SL. The significance of differences between group means was assessed by paired and nonpaired Student's f-tests.
COMPUTERIZED infrared
thermography (IRT) and magnetic resonance imaging (MRI) are useful in the assessment of low back pain. Thermography measures local skin temperature and therefore provides physiological data mediated through sympathetic nerve fibres. Characteristic skin surface temperature changes occur with lumbar nerve root irritation [1], non-radicular referred leg pain [2], and with lumbar disc pathology [3]. The diagnostic information shows significant correlation with CT scan, myelography [4] and MRI [5]. Magnetic resonance imaging provides fine anatomical details of the lumbar disc, nerve root and associated structures and can reveal changes which may have clinical relevance. In the present study, both visual and computerized IRT of patients with chronic lumbar pain with and without leg discomfort were compared prospectively with MRI and previously performed CT scans, myelography and discography. Many of the reported comparative studies between IRT, MRI and conventional radiological tests have relied largely on visual assessments of side-to-side leg temperature differences rather than more objective computer derived temperatures. PATIENTS AND METHODS Sixty-five patients with chronic back pain with and without referred leg pain (radicular or non-radicular) were assessed prospectively with MRI and IRT. Sixtyone of these were also examined by lumbar CT scan, 41 by myelography and 12 by discography. Thirty-seven were male and 28 were female, with a mean age of 41 (range 23-71 years). The mean duration of back pain at initial assessment was 2 years (range 2 months to 14 years). Thermographic assessments were performed with an Agema 782 infrared system using CATS (computer assisted thermographic software) and an IBM PXT for digitization of signals, storage and analysis of data. All patients followed a strict preparation protocol prior to IRT involving fasting for 3 h with no analgesics, smokSubmitted August 1989; revised version accepted 1 January 1990. Correspondence to Dr D. Thomas. 268
Downloaded from http://rheumatology.oxfordjournals.org/ at University of Iowa Libraries/Serials Acquisitions on May 29, 2015
SUMMARY Sixty-five cases of chronic low back pain were studied. Infrared thermography (IRT) was abnormal in 92%, magnetic resonance imaging (MRI) in 89%, computerized tomography (CT) in 87% and myelography in 80%. IRT correlated with MRI in 94% of cases, and with CTin87% of cases. Of 22 MRI positive disc and root cases, 21 (95%) had significant leg abnormalities on IRT. All 19 cases with radicular involvement on CT and all 18 with radicular involvement on myelography demonstrated significant leg changes on IRT.
THOMAS ETAL.: IRT, MRI, CT AND MYELOGRAPHY IN LOW BACK PAIN
l(a).
FIG.
l(b).
FIG.
3.
FIG. 1(C).
FIG.
2.
FIG. 1.—Thermography images, (a) Lumbar and gluteal showing a left lumbar flare, (b) Anterior leg (left cooler), (c) Plantar feet (left cooler). FIG. 2.—Magnetic resonance image of patient in Fig. 1. Transverse view at L4-5 level showing a left prolapse (arrowed) with extensive fibrosis extending from the theca laterally around the left L5 nerve root. FIG. 3.—Myelographic image of same patient. Left lateral obligue view. Largefillingdefect (arrowed) and non-filling of left L5 root.
Downloaded from http://rheumatology.oxfordjournals.org/ at University of Iowa Libraries/Serials Acquisitions on May 29, 2015
FIG.
269
270
BRITISH JOURNAL OF RHEUMATOLOGY VOL. XXIX NO. 4
OVERALL CHANGES FOR
TABLE I MRI, IRT, CT, MYELOGRAPHY
AND DlSCOGRAPHY
MRI (%)
IRT (%)
CT (%)
Myelography (%)
Discography (%)
Normal Abnormal Total
7 (11) 58 (89) 65 (100)
5 (8) 60 (92) 65 (100)
8 (13) 53 (87) 61 (100)
8 (20) 33 (80) 41(100)
1 (8) 11 (92) 12(100)
Abnormal* Disc only Disc & root Root only Total
36 (62) 22 (38)
12 (20) 32 (53) 16 (27) 60(100)
34 (64) 19 (36)
15 (45) 17 (52) 1 (3) -(100)
11(100) — — — — 11 (100)
58(100)
53(100)
* Lumbar and leg instead of disc and root for IRT group.
Four with bilateral referred leg pain and two with bilateral nerve involvement were excluded from subgroup analysis. Also excluded was a case with proven TB of the right sacroiliac joint and another with combined left sciatica and knee pain after a fracture. RESULTS Figures 1-3 illustrate a case of discogenic back pain and left sciatica after a discectomy at the L4-5 level, using imaging modalities employed in this study. Sixty-five cases of back pain were assessed by MRI and IRT and the results are summarized in Table I. The nerve root changes comprised loss of epidural fat, definite nerve root displacement or compression by postoperative fibrous tissue. Comparisons between MRI and IRT results are shown in Table II. An overall correlation between MRI and IRT was present in 89%. Lumbar CT scans were abnormal in 87% of cases (Table I). Comparisons between CT and IRT changes are shown in Table III. All cases with radicular involvement on CT show definite thermal asymmetry in the legs. There was an overall correlation between CT and IRT in 87%. Lumbar myelograms were abnormal in 80% of cases (Table I). Comparisons between myelography and IRT results are shown in Table IV. In all of the 18 cases where myelography demonstrated nerve root involvement, IRT demonstrated significant thermal asymmetry in the legs. Twelve also demonstrated lumbar changes on IRT. Twelve discograms were performed and 11 were abnormal (92%), showing leakage of dye with typical back pain (Table I). All were abnormal on IRT.
Objective analysis of thermal data For the purpose of more detailed computerized thermographic analysis, three well defined back pain groups were formed. Group 1, consisted of 21 cases with unilateral radiculopathy and discogenic back pain as defined by MRI and or CT and myelography; group 2, 12 with central discogenic back pain without unilateral or bilateral radiculopathy or symptomatic leg pain; group 3, 18 with lumbar disc degeneration and non-radicular referred leg pain. i.e. there was no nerve root involvement on MRI, CT and myelography. Thermal asymmetry The mean side-to-side temperature differences and standard deviation for the six specific thermographic regions for groups 1 and 2 is given in Table V and for group 3 in Table VI column 2. The mean temperature differences shown demonstrate that the symptomatic side of the body was cooler than the asymptomatic side. Side-to-side temperature difference in group 1 (unilateral radiculopathy cases) was maximal in the posterior leg region (—1.33°C) followed by the anterior leg region (-0.92°C) then the plantar foot region (—0.75°C). In group 2 (discogenic non-radicular nonreferred leg pain cases) the side-to-side temperature differences for all specific regions were negligible. Comparison between groups 1 and 2 demonstrated statistically significant differences for anterior leg, posterior leg, plantar foot, and anterior foreleg regions (Table V) and no significant differences for gluteal and anterior thigh areas. In group 3 (discogenic non-radicular referred leg pain cases) the side-to-side temperature differences were smaller than in group 2 but showed similar regional temperature distributions.
TABLE II
TABLE III
M R I COMPARED WITH I R T FOR DiSC AND NERVE ROOT CHANGES
C T COMPARED WITH IRT FOR DiSC AND NERVE ROOT CHANGES
Infrared thermographic abnormalities MRI changes
N
Disc only Disc & root Total
36 22 58
Lumbar only
Leg only
12 (33%) 8 (22%) 1 (5%) 6 (27%)
% relate to specific MRI subgroups.
Lumbar &leg
Infrared thermographic abnormalities
Normal
CT changes
N
15 (42%) 1 (3%) 15 (68%) 0
Disc only Disc & root Total
34 19 53
Lumbar only
Leg only
Lumbar &leg
Normal
9 (26%) 0
9 (26%) 6 (32%)
15 (44%) 13 (68%)
1 (3%) 0
% relate to specific CT subgroups.
Downloaded from http://rheumatology.oxfordjournals.org/ at University of Iowa Libraries/Serials Acquisitions on May 29, 2015
Overall changes
THOMAS ETAL.: IRT, MRI, CT AND MYELOGRAPHY IN LOW BACK PAIN TABLE IV MYELOGRAPHY COMPARED WITH IRT FOR DISC AND NERVE ROOT CHANGES
Infrared thermographic abnormalities N
Disc only Disc & root Root only Total
15 17 1 33
Lumbar only
Leg only
7(46%) 0 0
4(27%) 5(29%) 1 0
Lumbar &leg
Normal
4(27%) 12(71%)
— 0 0
% relate to specific myelography subgroups.
Comparisons between groups 1 and 3 demonstrated statistically significant differences for side-to-side temperatures in the posterior leg region only (P value 0.05
NS, not significant; Sign, significant; Af°C, mean symptomatic to asymptomatic side temperature difference.
TABLE VI GROUP 1 LUMBAR DISC DEGENERATION WITH UNILATERAL RADICULOPATHY (RADIC) (21 CASES) COMPARED WITH GROUP 3 LUMBAR DISC DEGENERATION AND NON-RADICULAR REFERRED LEG PAIN (REF. PAIN) 18 CASES MEAN SIDE-TO-SIDE TEMPERATURE DIFFERENCE
Radic
Ref. pain
Thermographic region
A/°C
(lSD)
At°C
(lSD)
(Test
Significance
Anterior leg Posterior leg Gluteal Plantar foot Anterior foreleg Anterior thigh
-0.92 -1.33 -0.18 -0.75 -0.47 -0.15
(0.92) (0.82) (0.33) (0.52) (0.49) (0.34)
-0.52 -0.56 -0.07 -0.43 -0.21 -0.06
(0.6) (1.19) (0.29) (0.74) (0.36) (0.32)
P>0.l P0.l
NS Sign NS NS NS NS
NS, not significant; Sign, significant; Ar°C, mean symptomatic to asymptomatic side temperature difference.
Downloaded from http://rheumatology.oxfordjournals.org/ at University of Iowa Libraries/Serials Acquisitions on May 29, 2015
Myelography changes
271
272
BRITISH JOURNAL OF RHEUMATOLOGY VOL. XXIX NO. 4 TABLE VII
GROUP 3 LUMBAR DISC DEGENERATION WITH NON-RADICULAR REFERRED LEG PAIN (REF. PAIN) 18 CASES, COMPARED WITH GROUP 2 LUMBAR DISC DEGENERATION WITHOUT RADICULOPATHY OR REFERRED LEG PAIN (NON-RADIC) 12 CASES. MEAN SIDE-TO-SIDE TEMPERATURE DIFFERENCES
Ref. pain
Non-radic
Ar°C
(lSD)
Ar°C
(lSD)
Anterior leg Posterior leg Gluteal Plantar foot Anterior foreleg Anterior thigh
-0.52 -0.56 -0.07 -0.43 -0.21 -0.06
(0.6) (1.19) (0.29) (0.74) (0.36) (0.32)
+0.10 -0.02 -0.10 +0.04 +0.05 +0.04
(0.22) (0.27) (0.23) (0.26) (0.25) (0.19)
f-Test
Significance
P
