GLOBUS MEDICAL, INC.
`EXHIBIT 1022
`IPR2015-to be assigned
`(Globus v. Bonutti)
`Page 1 of 7
`
`

`
`SAGITTAL LUMBAR SPINE
`
`381
`
`back pain subjects. However, studies using asymptomatic
`subjects as normal subjects are usually cross-sectional in
`design. More recently, the definition of what constitutes
`normal in asymptomatic subjects has received some criti-
`cism by Teresi et al. (25). They suggested that the desig-
`nation of asymptomatic is often only based on a question-
`naire or review of medical records, whereas a complete
`neurologic and/or physical examination is extremely rare
`because of limitations in time, money, and examiners.
`Besides the variations in normal subject groups, position-
`ing and measurement methods vary considerably.
`Studies of normal vertebral alignment in the sagittal plane
`have been conducted with inconsistent methodologies.
`These inconsistencies have resulted in data that, too fre-
`quently, are neither reproducible nor standardized, result-
`ing in limited agreement among the findings of investiga-
`tors. For example, variances include a mix of upright
`(2,l2,l5,16,23,28,30), sitting (1,29), and recumbent
`(6,8,9,28) radiographic positioning, whereas sample selec-
`tions range from asymptomatic (2,l2,l6,23,27,28,30) to
`symptomatic
`subjects
`(9,1 1,15,18,30)
`and living
`(2,l2,l6,23,28,30) versus deceased subjects (6,8). Also,
`there exist multiple methods of lordotic curve measurement
`procedures including Cobb angles at different levels,
`T11—Sl (15), T12—L5 (2,21), T12—S1 (12,27,28), L1—S1
`(l6,30), and L2—S1 (1 1), vertebral endplate angle to hori-
`zontal (6,8,9,2l), vertebral endplate angle to sacral base
`(24), and posterior vertebral body tangent lines (2,26). Fig-
`ure 1 illustrates some of these different methodologies.
`When comparing reported values of lumbar lordosis, the
`mixture of positioning and measurement methods creates
`some confusion. However, because the configuration of the
`lumbar lordosis is related to segmental loads and spinal
`coupling (19), normal values of lordosis would be clini-
`cally significant. We will show that standing lateral lum-
`bar lordotic values, derived from a reliable two-line mea-
`surement analysis, can be used to determine a standard
`lordotic alignment.
`In this study, measurement findings from lateral radi-
`ographs of the upright static lumbar spine in a normal pop-
`ulation are presented. The subjects had thorough physical,
`orthopedic, and neurologic examinations after a complete
`history. These findings are compared with and contrasted
`to other studies with similar radiographic positioning and
`measurement methodologies.
`The use of relative rotation angles (segmental) will be
`shown to be the ingredient that makes comparison and clar-
`ification of sagittal lumbar studies possible.
`
`METHODS
`
`The lateral lumbar spine radiographs of 50 consecutive
`subjects who fit the inclusion criteria described herein were
`
`*§€
`
`FIG. 1. Varying methods in radiographic analysis of lumbar lor-
`dosis. A: Some of the levels of Cobb angles. B: Segmental angles
`at inferior endplate compared to horizontal. C: Segmental angles
`at superior endplates add to global angle. D: Posterior body tan-
`gents can be segmental or global.
`
`retrospectively selected from the files of a central Illinois
`spine treatment center. The subjects comprising the study
`population were employee applicants for a local packing
`and shipping company who, as a condition of employment,
`were required to undergo a preemployment screening
`process including a thorough medical history, neuromus-
`culoskeletal examination, and lumbar spine radiographic
`examination. All examinations occurred between October
`
`1989 and May 1991.
`Inclusion criteria for the retrospectively selected sub-
`jects included: (a) no prior history of back trauma or dis-
`abling back pain (i.e., pain requiring restriction of occu-
`pational duties or avocational pursuits); (b) no history of
`treatment for any type of back or back-related condition;
`(c) normal neuromusculoskeletal examination results; and
`(d) no radiographic evidence of congenital anomaly, degen-
`erative joint disease, or other osseous disease.
`The lateral radiographs were taken at 40—in focal film
`distance on standard 14- x 17-in radiographic film. All
`subjects were standing in bare feet with the right side
`against the grid cabinet. Subjects were asked to assume a
`comfortable position with the hands at the sides. The sub-
`jects were then instructed to place the hands on top of the
`
`J Spinal Dimrzl. Vol. I0. Na. 5. I997
`
`Page 2 of 7
`Page 2 of 7
`
`

`
`382
`
`S. J. TROYANOVICH ETAL.
`
`head with elbows in the coronal plane of the body. This
`standardized hand and arm placement was used to main-
`tain the weight of the upper extremities as close as possi-
`ble to the midaxillary line, as in neutral, comfortable, erect
`posture. The patient was then instructed to move to their
`right until their hips or rib cage contacted the grid cabinet.
`Tangent lines were drawn at the posterior vertebral mar-
`gins of T12 through S 1. Angular positions of adjacent ver-
`tebrae (T12—Ll, L1—L2, L2-L3, L3—L4, L4-L5, L5-S1),
`referred to as relative rotation angles (RRA), and the angu-
`lar positions of L1 compared with L5, termed ARA, were
`used as segmental measurements and overall measurement
`of the degree of lumbar lordosis, respectively. Also mea-
`sured were a Cobb angle at T12 to S 1, Ferguson’s angle
`(sacral base relative to horizontal), and a vertical sagittal
`alignment of the perpendicular distance from the pos-
`teroinferior vertebral margin of S1 to a vertical axis line,
`
`which was drawn inferiorly and parallel to the lateral edge
`of the radiographic film from the posteroinferior vertebral
`body margin of T12. The angle of pelvic tilt, which was
`drawn as a line extending from the uppermost portion of
`the femur—acetabulum joint to the posterior-inferior ver-
`tebral margin of S1, was compared with a true horizontal
`line. Figure 2 illustrates all these lines and angles. The radi-
`ographic analysis procedure described by Troyanovich et
`al. (26) was used to analyze the films. This method of
`analysis, using a two-line analysis method, was selected
`for use because of its high degree of reliability and low
`standard error of measurement, averaging ~l.5° (26). This
`is in contrast to Cobb angles, which require a four-line
`analysis. As examples, a 1988 study (14) on AP thoracic
`Cobb angles reported a minimum variability of 2.5°, and
`a recent study of the reliability of deriving Cobb angles in
`the sagittal plane reported variation of repeat measures as
`high as 10° (20).
`
`TANGENTS
`
`RESULTS
`
`The means, standard deviations, and minimum and max-
`imum values for each of the radiographic measures, age,
`height, and weight for each of the subjects are reported in
`Table 1. The mean age was 27.1 years, with a range of
`17-52 years of age, and a standard deviation of 8 years.
`For the physical characteristics, the mean height was 171
`cm (67.5 in), with a range of 152-188 cm (60-74 in) and
`a standard deviation of 9 cm (3.6 in). The mean weight
`was 70.9 kg (156.2 lb), with a range of 49.5—1 13.5 kg
`(109—250 lb) and a standard deviation of 15.4 kg (34 lb).
`To briefly summarize the average angle magnitudes in
`Table 1, for the ARA between L1 and L5 and the RRA at
`each interspace, we have ARALLL5 = —39.7°, RRAT12_L,
`= -3.60, RRALl_Lz = "4.1°, RRAL2_L3 = -7.60,
`
`TABLE 1. Statistical radiographic measures in 50 subjects
`without history of low back injury, low back pain, degenerative
`changes, or congenital spinal zmomalies
`
`Category
`
`Mean
`
`Age (yrs)
`Height (cm)
`Weight (kg)
`Ll—L5 ARA* (°)
`Tl2—Ll RRA‘ (°)
`L1-L2 RRA (°)
`L2-L3 RRA (°)
`L3-L4 RRA (°)
`L4-L5 RRA (°)
`L5-S1 RRA (°)
`Ferguson (°)
`Pelvic tilt (°)
`T1 ++(mm)
`
`27.1
`171.5
`70.9
`-39.7
`-3.6
`-4.]
`-7.6
`-11.7
`-16.8
`-32.4
`39.2
`48.9
`-5.7
`
`SD
`
`8.0
`9.1
`15.4
`9.1
`2.6
`3.0
`2.8
`2.8
`4.5
`7.9
`6.5
`8.6
`17.3
`
`Minimum Maximum
`
`17.0
`152.4
`49.5
`-60.5
`-10.0
`-10.5
`-14.0
`-19.3
`-27.0
`-47.5
`26.0
`26.7
`-48
`
`52.0
`188.0
`113.5
`-22.5
`0.0
`0.0
`-1.3
`-6.0
`-7.8
`-14.5
`49.5
`69.]
`35
`
`ARA*, absolute rotation angle; RRA‘, relative rotation angle; '1‘Z++,
`sagittal vertical alignment of T12-S I.
`
`Page 3 of 7
`Page 3 of 7
`
`
`
`FERGUSON'S
`ANGLE
`
`ANGLE op
`PELVIC TILT
`
` HORIZONTAL
`
`FIG. 2. Fladiographic lines, angles. and distance used in analy-
`sis of lordosis. Tangent lines are used lor intersagmental relative
`rotation angles (RRA) and global lordosis (ARA) at L1-L5. Fer-
`guson's angle was measured as sacral base relative to a true hor-
`izontal line. The angle of pelvic tilt was drawn from the uppermost
`portion of the acetabulum to the posterior—inferior vertebral mar-
`gin of S1. Translation of vertical alignment of T12 ls measured at
`the posterior—inferior body corner of 81.
`
`J Spinal Disartl. Vol. ID. No. 5, 1997
`
`S1
`
`TRANSLATION
`DISTANCE TO
`POSTERIOR-INFERIOR
`BODY MARGIN OF SI
`
`ARCUATE
`LINE
`
`
`
` HORIZONTAL
`
`

`
`SA GITTAL LUMBAR SPINE
`
`383
`
`= -11.7°, RRA,_,,_L5 = —l6.8°, and RRAL5_s, = -32.4°. The
`Ferguson sacral base angle average was 39.2°, whereas the
`mean angle of pelvic tilt was arcuate = 48.9°. The average
`vertical translation of the inferior-posterior body corner
`of T12 compared with the inferior-posterior corner of S1
`was posterior, —Tz = 5.7 mm. It is noted that the relative
`rotation angles increase in magnitude from T12 to S1.
`Table 2 lists the means, standard deviations, and ranges
`for height, weight, and each of the radiographic measures
`for the 21 female and 29 male subjects separately. Little
`difference exists between the sexes. For example, all the
`angles for males and females differ by 2° or less except the
`difference in the angle of pelvic tilt (arcuate) is 3.8°. The
`females had a larger posterior translation of T12 to S1.
`However, the female average weight was 156 lb, which
`suggests that the difference in pelvic tilt and posterior trans-
`lation might be due to this average weight being above the
`norm.
`
`DISCUSSION
`
`Clinicians have long held the notion, from years of empir-
`ical observation without sufficient data (3,5,22), that exces-
`sive lordosis is a causative factor in low back pain. In efforts
`to investigate this hypothesis, researchers have attempted
`to study the geometric configuration of the sagittal lumbar
`spine using a variety of radiographic positioning and men-
`suration schemes. To varying degrees, design differences
`from one study to another have complicated the issue, mak-
`ing it difficult to draw any definitive conclusions regard-
`ing lumbar lordosis and its possible relationship to lumbar
`pain or spinal dysfunction syndromes. These variances
`include a mix of upright and recumbent radiographic posi-
`tioning procedures, multiple methods of lordotic curve mea-
`surement procedures, and sample selections ranging from
`
`asymptomatic to symptomatic subjects, and living versus
`deceased subjects. These inconsistencies in lumbar curve
`assessment have led to a situation in which it is commonly
`accepted in the health care sciences that there exists no
`species-specific norm unique to the sagittal plane orienta-
`tion of the lumbar spine, but only individual norms (17).
`We suggest a different perspective. First, it is noted that
`our asymptomatic group meets the criteria of comprehen-
`sive examinations and histories suggested by Teresi et al.
`(25) for normal groups. Second, from the multitude of stud-
`ies on lumbar lordosis, six studies have been chosen to be
`compared with our study. These sagittal lumbar studies
`(2,l2,l6,24,28,30) have been undertaken to evaluate lum-
`bar lordosis in asymptomatic populations in whom a rel-
`atively standardized weight-bearing (standing) radiographic
`positioning procedure was used. Because of the variable
`methods used to derive measures of spinal configuration,
`direct comparisons of the findings are difficult but not
`impossible. The specific findings of each of these studies,
`including the present study, are summarized in Table 3.
`An important observation in Table 3 is that segmental con-
`tributions and overall lumbar lordosis are very similar in
`these seven studies, even though the average ages of the
`subjects are 13, 18, 24, 27, 39, 40, and 57 years, i.e., a wide
`range of ages. These values are discussed further.
`In a study performed by Stagnara et al. (24), the upright
`radiographs of 100 young adults (ages 20-29 years) with no
`known spinal disease were examined to detemiine the max-
`imum ranges of thoracic kyphosis and lumbar lordosis in this
`population. They reported the overall mean lordosis as deter-
`mined by vertebral body endplate inclination analysis as
`CobbT12_s, = -55“. In their Table 4, Stagnara et al. (24) report
`the individual vertebral segmental contributions to lordosis,
`RRAT12-L1 = +1°v RRAL1-L2 = ‘2°s RRAL2-L3 = “7°s RRA1.3-
`L4 = -11°, RRAL4_,_5 = —l5°, and RRA,_5_S1= -21°.
`
`TABLE 2. Statistical rangesfor rarliograplzic nzeasures for males versus females
`Mean
`SD
`Minimum
`
`Maximum
`
`Category
`
`Age (yrs)
`Height (cm)
`Weight (kg)
`L1-L5 ARA* (°)
`T12-L1 RRA‘ (°)
`1.1-1.2 RRA (°)
`L2—L3 RRA (°)
`L3—L4 RRA (°)
`L4-L5 RRA (°)
`L5-S1 RRA (°)
`Ferguson (°)
`Pelvic tilt (°)
`T1 (mm)
`
`M
`
`25.8
`176.3
`71.0
`-40.7
`-3.5
`-4.2
`-7.6
`-11.7
`-17.3
`-31.9
`39.9
`47.3
`-2.5
`
`F
`
`28.6
`164.6
`70.9
`-38.3
`-3.8
`-3.9
`-7.6
`-11.6
`-16.0
`-33.1
`38.2
`51.1
`-9.3
`
`M
`
`8.1
`6.1
`13.9
`9.9
`2.6
`3.1
`2.7
`2.8
`4.8
`7.2
`6.7
`7.5
`14.8
`
`F
`
`8.0
`8.4
`17.7
`7.9
`2.7
`2.8
`2.9
`2.9
`3.9
`8.9
`6.2
`3.7
`20.2
`
`M
`
`17.0
`165.1
`53.6
`-60.5
`-10.0
`-10.5
`-12.5
`-19.3
`-27.0
`-44.8
`28.0
`26.7
`-29.2
`
`F
`
`18.0
`152.4
`49.5
`-50.3
`-10.0
`-10.5
`-14.0
`-16.3
`-24.0
`-47.5
`26.0
`36.8
`-48.0
`
`M
`
`52.0
`188.0
`110.3
`-26.5
`0.0
`0.0
`-1.3
`-6.0
`-7.8
`-14.5
`49.0
`69.1
`31.4
`
`F
`
`41.0
`180.3
`113.5
`-22.5
`0.0
`0.0
`-3.3
`-6.5
`-9.0
`-19.5
`49.5
`64.3
`35.0
`
`ARA*, absolute rotation angle; RRA‘, relative rotation angle; ++Tz, sagittal vertical alignment of T12-S1.
`
`J Spinal Disord. Vol. 10, Na. 5. I997
`
`Page 4 of 7
`Page 4 of 7
`
`

`
`384
`
`S. J. TROYANOVICH ETAL.
`
`TABLE 3. Comparisons with six prior studies of the lumbar [0I‘t10.S‘iS in the standing position
`
`Stagnara
`Wambolt
`Bernhardt
`Jackson
`Gelb
`Wood
`Current
`
`Category
`et al. (24)
`et al. (28)
`Bridwell (2)
`McManus (16)
`et al. (12)
`et al. (30)
`study
`
`Subjects
`Mean age
`Method of measure
`
`43 F, 57 M
`24.5
`Cobb
`
`28 F, 22 M
`18.0
`Cobb
`
`Overall lordosis
`
`infTl2-sup S1
`
`infTl2—sup Sl
`
`55 F, 47 M
`12.8
`Cobb and Post
`tangents
`1nfTI2—inf L5
`
`50 F, 50 M
`38.9
`Cobb
`
`54 F, 46 M
`57.0
`Cobb
`
`28 F, 22 M
`40.1
`Cobb
`
`sup Ll-sup S1
`
`infTl2-sup S1
`
`sup L1 to sup SI
`
`21 F, 29 M
`27.1
`Post tangents
`and Cobb
`ARA Ll—5,
`Cobb T12—S1
`
`-39.7”, —65°
`—58.8°
`—64°
`-60.9°
`-44°
`-59°
`—55°
`Degrees
`—3.6°
`+2.1°
`+2°
`(unreported)
`+l°
`—3°
`+l°
`RRA+ Tl2—Ll
`-4. 1°
`—l.5°
`-4°
`-1.7“
`-4°
`—3°
`-2°
`RRA Ll—2
`-7.6”
`-7.4”
`-10”
`—7°
`—7°
`-6°
`-7°
`RRA L2—3
`-11.7"
`-11“
`—l4°
`—11.3°
`-13“
`—ll°
`-11°
`RRA L3—4
`-l6.8°
`-16.l°
`-24°"
`—16.5°
`-20°
`-15°
`-15°
`RRA L4—5
`—25.4°”
`—22.3°
`—24°
`-24.6”
`-28“ + 7° = -21°”
`-21“
`—2l°
`RRA L5—S1
`
`
`
`
`
`
`
`—55° -59“ -65°” —60.9° -5401: -56.7°Cobb Tl 2—Sl -65”
`
`F, female, M. male; ARA, absolute rotation angle; RRA, relative rotation angle.
`"Sum of RRAs does not equal the reported Cobb angle, suggesting an error.
`”Ang|e between posterior tangents corrected by adding 7°.
`
`Wambolt and Spencer (28) describe the findings of their
`study of the upright radiographs of 50 young “normal”
`individuals (average age 18 years). They describe the over-
`all average lordosis as determined from constructing Cobb
`angles from the inferior endplates of T12 and the superior
`endplates of S1 as -59°. They go on to report separately
`the individual contributions to the lordotic curvature from
`
`both the vertebral bodies and the discs. By adding their
`reported values for the intervertebral disc with its superior
`vertebral body, the relative segmental contributions to over-
`all lordosis from Tl2—Ll through L4—L5 are -3°, —3°,
`—6°, -11°, -15°, and -21°, respectively.
`Bernhardt and Bridwell (2) reported the values deter-
`mined from analyzing the upright radiographs of 102 young
`healthy individuals (average age 12.8 years). They mea-
`sured the overall lordosis by constructing Cobb angles from
`the inferior endplates of T12 and L5. Their reported over-
`all mean lordosis for this measure was —44°. They reported
`relative intersegmental contributions to lumbar lordosis by
`constructing lines across the posterior vertebral body mar-
`gins of each adjacent pair of vertebrae and measuring the
`angle of intersection formed by these lines. This is identi-
`cal to the method described in the radiographic measura-
`tion reliability study used by us (26). The values derived
`from this method were +1, -4, -7, -13, -20, and —28° from
`T12-Ll through L5—S1, respectively.
`Jackson and McManus (16) described a similar study
`of 100 upright radiographs taken from asymptomatic vol-
`unteers ages 20-65 years with no prior history of lumbar
`spine surgery or spinal deformity. They reported the over-
`all lumbar lordosis as a Cobb angle of —60.9° derived from
`the superior endplates of L1 and S1. They then went on to
`report Cobb angles of adjacent vertebrae as Ll—L2 = —l.7°,
`L2—L3 = -7.0°, L3-L4 = —11.3°, L4-L5 = —16.5°, and
`
`J Spinal Disartl, Vol. /0, No. 5, I997
`
`L5—S1 = -24.6°. In this study, however, their results were
`compared with the upright radiographs of 100 low back
`pain patients who were matched for age, sex, and size.
`They reported a significantly lesser magnitude of lumbar
`lordosis in the patient group versus the volunteer group.
`They further stated that in terms of the segmental contri-
`butions to the sagittal curve, on average, the patient group
`tended to have more upper lumbar lordosis and less lower
`lumbar lordosis than did the volunteer group, leading them
`to conclude that, “ .
`.
`. a loss of segmental lordosis in the
`lower lumbar area maybe [sic] possibly associated with
`low back pain” (16).
`Gelb et al. (12) described their findings in 100 asymp-
`tomatic middle-aged and older volunteers (mean age 57
`years). They reported their findings from upright radi-
`ographs by determining the Cobb angle from the inferior
`endplate of T12 and the superior endplate of S1. The sub-
`jects’ overall mean total lordosis was CobbT,2_s] = -64°.
`The segmental contributions were reported by Gelb et al.
`as +2° for Tl2—Ll , —4° for Ll-L2, -10° for L2—L3, -14°
`for L3-L4, -24° for L4-L5, and -24 for L5—S 1.
`Finally, Wood et al. (30) evaluated 50 upright radi-
`ographs taken from asymptomatic volunteers ages 26-59
`years with no known spinal deformity, no prior history of
`spinal fracture, no prior history of spinal tumor or infec-
`tion, and no prior history of lumbar spine surgery. They
`reported the overall lumbar lordosis as a Cobb angle of
`—58.8° derived from the superior endplates of L1 and S 1.
`They then went on to report Cobb angles of adjacent ver-
`tebrae as Tl2—Ll = +2.1°, Ll—L2 = —l.5°, L2—L3 = —7.4°,
`L3-L4 = —1l.0°, L4—L5 = -l6.l°, and L5—S1 = —22.3°.
`As with the Jackson and McManus (16) study, Wood and
`colleagues compared their results with the upright radi-
`ographs of 50 additional low back pain patients who were
`
`Page 5 of 7
`Page 5 of 7
`
`

`
`SA GITTAL LUMBAR SPINE
`
`385
`
`closely matched for age, sex, and size. They reported a
`slightly lesser magnitude of overall lumbar lordosis in the
`patient group versus the volunteer group. A further analy-
`sis of their data indicate that in terms of the segmental con-
`tributions to the sagittal curve, the patient group tended to
`have more lordosis at Ll—L2 and less lumbar lordosis at
`
`L4-—L5 and L5—S1. This finding would be consistent with
`those reported in the patient group of Jackson and
`McManus (16).
`Because of the varying methods used to measure the
`lumbar lordosis and the relative contributions made at each
`
`interspace, few direct comparisons can be made between
`these studies. From Table 3, the overall measurements of
`lordosis by Wambolt and Spencer (28) and Gelb et al. (12)
`are identical methods, and their mean values are only 5°
`apart in Cobb angle at T12—S1. Furthermore, the methods
`used by Jackson and McManus (16) and Wood et al. (30)
`are identical. Their data are likewise nearly identical, with
`an average difference of 0.72° per interspace. Addition-
`ally, the intersegmental values for Tl2—Ll through L5—S1
`determined by Bernhardt and Bridwell (2) and in the cur-
`rent study are also derived in an identical manner. Com-
`parison of these values demonstrates extremely close mag-
`nitudes with an average difference of 0.87° per interspace.
`A general trend can be seen when looking at the over-
`all data sets from each of these studies. It seems clear that
`
`the usual, typical, or average morphology for the sagittal
`lumbar spine is that of lordosis beginning at ~T12 or L1.
`The segmental contribution to the lordotic curve increases
`distally, with the greatest contribution coming from the
`two most caudal segments. Depending on the study cited,
`the two most distal segments contribute between 54 and
`68% of the total lordotic curve. Our data tend to lend sup-
`port to this hypothesis as well. A very important observa-
`tion in Table 3 is that all the average relative rotation angles
`for a particular segmental level are nearly the same, with
`an absolute value maximum range of 2—5°, in these seven
`different investigations. This suggests the existence of a
`normal (ideal) lumbar lordosis.
`For future studies on spinal sagittal curves, the authors
`recommend that researchers provide all relative rotation
`angles (RRA) with their angle for measuring total lordo-
`sis or kyphosis (ARA). This will provide the opportunity
`for comparison of results among different studies. For
`example, if Tl2—Sl, T12—L5, Ll—L5, or L1—Sl is given
`as the total lordosis with all RRAs provided, then angles
`can be added or subtracted from each method to arrive at
`
`an accurate comparison. By adding our Tl2—Ll RRA to
`our ARA from L1—L5, the authors obtain —3.6° + —39.7°
`
`= -43.3”, which is very close to the ARA of —44° reported
`by Bernhardt and Bridwell (2).
`It is noted that the contribution of the first sacral seg-
`ment to lordosis depends on whether lines are drawn on
`
`the posterior body of S1 or on the superior surface of S 1.
`This is due to the increased angulation of the sacral base
`from perpendicular to the posterior body margin. If one
`adds —39.7° to the L5—S1 angle of -32.4“ and subtracts the
`difference of the angle of the surface of the sacral base
`from perpendicular to the posterior body of S1 (—7°), the
`authors’ value, Cobb]-ms, = (—39.7° + —32.4°) — (— 7°) =
`—65°, is very close to the values of —59.0° reported by
`Wambolt and Spencer (28), —60.9° reported by Jackson
`and McManus (16), and 64° reported by Gelb et al. (12)
`for a lordosis value from inferior of T12 to the superior
`endplate of S1. Also, the same procedure corrects the error
`between the posterior tangent lines for Bernhardt and Brid-
`well (2), CobbT12_S1 = (—44° + —28°) — (—7°) = —65°.
`With the CobbTl2_S1 = —55° suggested by Stagnara et
`al. (24), the range of average CobbT,2_S1 angles in seven
`studies using the standing positioning in 552 normal sub-
`jects is —55° to —65°, with a mean of —61°. Using superior
`endplate lines, the segmental angles have surprising uni-
`formity in Table 3. Eliminating the unreported Tl2—Ll
`angle (16), the possible erroneous L4—L5 angle (12), and
`the 7° difference between perpendicular at S1 for the pos-
`terior tangent versus superior sacral base, the following
`averages from Table 3 are obtained: RRA-mi, = —O.1°,
`RRALl_L2 = -2.90, RRAL2_L3 = "7.4°, RRAL3_LA = "1 1.90,
`RRA[A_L5 = —'16.6D, and RRAL5_Sl = -22.80.
`Unlike the variables in the seated and recumbent pos-
`tures, the standing lateral position (posture) has been
`reported to be very repeatable over time, i.e., 2 years (4).
`Investigators (12,l5) have suggested that lumbar lordosis,
`angle of pelvic tilt, Ferguson's sacral base angle, and sagit-
`tal alignment of T12 compared with the sacrum, are func-
`tions of the postural position of the head, neck, and tho-
`racic cage above and the alignment of the lower extremities
`below. This suggests that the large range and standard devi-
`ation for the z-axis translation measurements of vertical
`
`alignment of T12 to S1 in Table 1 are due to a multitude
`of different full-body sagittal postures.
`
`CONCLUSION
`
`For practitioners who use anatomic outcomes, norma-
`tive data are necessary to establish meaningful clinical
`goals. Results from our study confirm and clarify similar
`results reported by others and indicate that the overall lum-
`bar lordotic curve is comprised of the sum of varying indi-
`vidual contributions from Tl2—Ll vertebrae to and inclu-
`
`sive of the L5—S1 disc space. These segmental contributions
`(RRAs) were found to be very similar at each level in stand-
`ing studies of lumbar lordosis, and these RRAs provide
`the means to compare and correlate the results of prior
`studies using different methodologies. The general trend
`indicates that there is minimal lordosis in the upper lum-
`
`J Spinal Disarrl, Vol. I0, N0. 5, I997
`
`Page 6 of 7
`Page 6 of 7
`
`

`
`386
`
`S. J. TROYANOVICH ET/lL.
`
`bar spine with increasing percentage, -65%, of lordosis dis-
`tally. The range of average Cobb angles (T12—Sl) in seven
`studies was -55 to -65°, with a mean of —6l°, whereas the
`segmental rotation angles averaged: RRAT,2_L, = -0.1°,
`RRAL]_L2 = “2.9°, RRAL2_L3 = “'7.4°, RRAL3_LA = -11.90,
`RRAL4_L5 = "16.6°, and RRAL5_Sl = -22.80.
`We have identified two major areas, which may help to
`minimize the confusion in comparing various studies. The
`use of relative rotation angles in any sagittal studies of the
`human spine and the additional measurement of the subjects’
`lateral degrees of freedom of the global parts of human pos-
`ture could be the variables needed to unite the diversity of
`published values for sagittal postural measurements.
`We believe there is significance in that 552 normal sub-
`jects, studied in seven different investigations with widely
`differing age groups, had very similar RRAs, ARAs, and
`Cobb angles in the sagittal lumbar view. Additionally, two
`of the prior studies discussed compared the results obtained
`from their asymptomatic volunteers with back pain patients
`and found differences between the geometric configura-
`tion of the upright asymptomatic lumbar spine and the geo-
`metric configuration of the painful upright lumbar spine.
`From a strictly mechanical standpoint, this is suggestive
`to us that there may exist an ideal sagittal lumbar curva-
`ture that may tend to protect holders of this geometric con-
`figuration against nociceptive tendencies.
`The increased curvature at L4—L5—S1 suggests that an
`elliptical model could approximate the ideal normal erect
`lumbar lordosis. In future studies, it is our intention to
`
`develop a geometric model of the ideal upright static lum-
`bosacral spine based on a partial portion of an ellipse.
`
`Acknowledgment: We thank Sanghak 0. Harrison, D.C., for
`artwork and CBP Non-profit, Inc., for support.
`
`REFERENCES
`
`l. Andersson GBJ, Murphy RW, Ortengren R, Nachemson A: The
`influence of backrest inclination and lumbar support on lumbar lor-
`dosis. Spine 4:52-58, 1979.
`2. Bernhardt M, Bridwell KH: Segmental analysis of the sagittal plane
`alignment of the normal thoracic and lumbar spines and thora-
`columbarjunetion. Spine 14:717—721, 1989.
`3. Bourdillon JF, Day EA, Bookhout MR: Spinal Manipulation, 5th
`ed. Oxford: Butterworth-Heinemann Ltd., 1992, p 77.
`4. Bullock-Saxton J: Postural alignment in standing: A repeatable study.
`Anst Pltysiotlter 39:25-29, 1993.
`5. Cailliet R. Low Back Pain Syndrome, 3rd ed. Philadelphia, FA Davis,
`1981, p 53.
`6. Cholewicki J, Crisco JJ, Oxland TR, Yamamoto I, Panjabi MM:
`Effects of posture and structure on three-dimensional coupled rota-
`tions in the lumbar spine. Spine 21:242l-2428, 1996.
`7. Deyo RA, Andersson G, Bombardier C, et al.: Outcome measures for
`studying patients with low back pain. Spine 19:2032s—2036s, 1994.
`
`8. Farfan HF, Haberdeau RM, Dubow HI: Lumbar intervertebral disc
`degeneration: The influence of geometrical features on the pattern
`of disc degeneration—A post mortem study. J Bone Joint Surg
`54:492—510, 1972.
`9. Fernand R, Fow DE: Evaluation of lumbar lordosis: A prospective
`and retrospective study. Spine 10:799-803. 1985.
`10. Fricton JR, Awad EA: Myofascial Pain and Fibromyalgia. New
`York, Raven Press, 1990.
`l 1. Frymoyer JW, Newburg A, Pope MH, Wilder DG, Clements J,
`MacPherson B: Spine radiographs in patients with low-back pain. J
`Bone Joint Surg [Am] 66: 1048-1055, 1984.
`12. Gelb DE, Lenke LG, Bridwell KH, Blanke KB, McEnery KW: An
`analysis of sagittal spinal alignment in 100 asymptomatic middle
`and older aged volunteers. Spine 20: 135 l-1358, 1995.
`13. Goel VK, Weinstein JN: Biomecltanics of the Spine: Clinical and
`Surgical Perspective. Boca Raton, Florida, CRC Press, 1990.
`14. Goldberg MS, Poittras B, Mayo NE, Labelle H, Bourassa R, Cloutier
`R: Observer variation in assessing spinal curvature and skeletal devel-
`opment in adolescent idiopathic scoliosis. Spine 1221371-1377, 1988.
`15. ltoi E: Roentgenographic analysis of posture in spinal osteoporotics.
`Spine 16:750-756. 1991.
`16. Jackson RP, McManus AC: Radiographic analysis of sagittal plane
`alignment and balance in standing volunteers and patients with low
`back pain matched for age, sex, and size. Spine 19:161 1-1618, 1994.
`17. Lewit K: Manipulative Therapy in Rehabilitation ofthe Locotnotor
`System, 2nd ed. Oxford, Butterworth-Heinemann. 1991, p 41.
`18. Nissinen M, Heliovaara M, Seitsamo J, Alaranta H, Poussa M:
`Anthropometric measurements and the incidence of low back pain
`in a cohort of pubertal children. Spine 19: 1367-1370, 1994.
`19. Cholewicki J, Crisc JJ, Oxland TR, Yamamoto I, Panjabi MM:
`Effects of postureand structure on three-dimensional coupled rota-
`tions in the lumbar spine. Spine 212421-2428, 1996.
`20. Polly DW, Kilkelly FX, McHale KA, Asplund LM, Mulligan M,
`Chang AS: Measurement of lumbar lordosis: Evaluation of intraob-
`server, interobserver, and technique variability. Spine 21: 1530-1536,
`1996.
`21. Pope MH, Bevins T, Wilder DG, Frymoyer JW: The relationship
`between anthropometric, postural, muscular, and mobility charac-
`teristics of males ages 18-55. Spine l0:644—643, 1985.
`22. Reilly B: Practical Strategies in Outpatient Medicine, Philadelphia,
`WB Saunders, 1984, p 6.
`23. Splithoff GA: Lumbosacral junction: Roentgenographic compar-
`isons of patients with and without backache. JAMA 152: 1610, 1953.
`24. Stagnara P, De Mauroy JC, Dran G, Fonon GP, Costanzo G, Dim-
`net .1, Pasquet A: Reciprocal angulation of vertebral bodies in a sagit-
`tal plane: Approach to references for the evaluation of kyphosis and
`lordosis. Spine 7:335—342, 1982.
`25. Teresi L, Lufkin RB, Reicher MA, et al.: Asymptomatic degenera-
`tive disk disease and spondylosis of the cervical spine: MR imag-
`ing. Ratliology 164:83-88, 1987.
`26. Troyanovich S, Robertson GA, Harrison DD, Holland B: lntra- and
`interexaminer reliability of the chiropractic biophysics lateral lum-
`bar radiographic mensuration procedure. J Manip Pltysiol Tlter
`182519-524, 1995.
`27. Voutsinas SA, MacEwen D: Sagittal profiles of the spine. Clin 0I‘llt0[)
`210:235-242, 1986.
`28. Wambolt A, Spencer DL: A segmental analysis of the distribution
`of the lumbar lordosis in the normal spine. Ortliop Transactions
`11:92-93, 1987.
`29. Williams MM, Hawley JA, McKenzie RA. van Wijmen PM: A com-
`parison of the effects of two sitting postures on back and referred
`pain. Spine 16:1 185-1 191, I991.
`30. Wood KB, Kos P. Schendel M, Persson K: Effect of patient posi-
`tion on the sagittal-plane profile of the thoracolumbar spine. J Spinal
`DlS0l'(l 9: 165-169, 1996.
`
`J Spinal Disortl, Vol. /0, No. 5, I997
`
`Page 7 of 7
`Page 7 of 7