335

Prlnted ln the U.S.A. Rep,inud /ru", Copyrighl r 1982 by Herper end Row Publlehere, Ine. SPI NE • VOLUME 7 • NUMBER 4 • 1982

Reciprocal Angulation of Vertebral Bodi'es in a

Sagittal IPlane: Approach to References for the

Evaluation of Kyphosis and Lo'rdosis

PEIRRE STAGNARA,* JEAN CLAUDE DE MAUROY,* GEORGES DRAN,* GEORGES P. GONON,t GIUSEPPE

COSTANZO,:t: JOANNES DIMNET,§ and ANNICK PASOUET§

This report establishes a table of references for kyphosis and lordosis since, to the best of the authors' knowledge, no such table exists. A sample of healthy individuals was selected and, by means of roentgen­ographic study, the authors were able to def,ine the reciprocal angula­tion of each vertebral body. The average values of the data obtained should provide adirectly applicable table of references, but four years of work in this direction has shown that there ,is no simple answer. However, this study does establish an easily reproducible reference position, criteria sample selection, roentgenographic and clinical measuring processes, and a method to computerize the information obtained. [Key words: kyphosis, lordosis, quantitation, reciprocal relationship ],

S ITTING AND SUPINE positions are extremely The clinical measurements are taken relative to the variable, but we have noticed that each vertical represented by a plumb line. (These measure­individual has a habituai standing posture. ments will not be given here.) In his thesis, Georges For subjects undergoing c1inical and X­ Dran studied the correlations between X-ray and clini­

ray examinations at intervals of five to ten years, and cal measurements, which are close enough to confirm where no growth or pathologie deformation factors are each other. to be taken into account, the c1inical and X-ray mea­ For a lateral roentgenogram of the spine in the surements of kyphoses and lordoses are remarkably standing position defined above, the humeri must be constant to within a few degrees, provided the position moved forward without modifying kyphosis or \ordo­is clearly stipulated. sis. We found lhat having the hands resling on a

This standing position, easily reproducible, is as horizontal support in front of the patient at pelvis level follows: legs straight and trunk and shoulder-girdle in did not change the clinical measurements, whereas rel~xed p sition (Figure 1). having the arms straight out in front or raised above

the head changed kyphosis and lordosis considerably. From the • Centre des Massues, Lyon, France, t Pav. T. Hopital

Ed. Herriot, Lyon, France, *C1inica Ortopedica dell'Universita, COMPOSING THE SAMPLE R ma, Itafia, . Ecole Centrale de Lyon, Ecully, France. Pre ented at the l5th Annual Meeting of the S oliosi Research For the first sample, we chose young adults with no S ciety. Chicago, Illinois, September 17-19, 1980.

ubmitted ~ r publication March 10, 1981. known spinal pathology, aged from 20 ta 29, inclusive­0362-243 /82/0922-0335$0120 @1982 Harper & Row. Pubtishen;, Inc.

36

SPINE • VOLUME 7 • NUMBER 4 • 1982

336 RECIPROCAL ANGULATION • STAGNARA ET AL

Fig 1. Lateral x-ray position.

Iy. This period of life appears to be probably the most stable; skeletal maturity has been reached and the vertebral modifications more frequent in older sub­jects have not yet appeared. 2 Our study particularly concerns a group of roo subjects from Lyons and Marseilles: 43 women and 57 men. A second check group of 37 subjects (19 women and 18 men) was made up of subjects from Giuseppe Costanzo.

Rome who were selected by

COMPUTERIZING THE INfORMATION

Digitalizer

The measurements were established by treating each roentgenogram on a digitalizing table. with four points defining each vertebral body. The bottom edge of the film was used as a reference to the horizontal. The program fed into the computer gives ail the reciprocal angulations defined above. We found the average value of each reciprocal angulation for each level, the standard deviations, and the upper and lower limits of the sample.

Thus, we were able to compose a table with two en tries and 15 columns with the values rounded off to the nearest degree. Our first observation is that. even for tbis small sample, the differences from one person

to the next are so great that average values cannot by any means be used as normative values (Table 1). Taking as an example the reciprocal angle of T Il with L3, the average value is - 5° (Iordosis 5°), but our sample includes ail ·intermediate values between a lordosis of - 30° and a kyphosis of + 15°, giving a 45° range (the standard deviation at this level being 9°).

Consequently, the average values can be used for a certain number of cases studied statistically, but for individual cases, it is better to use the upper or lowcr limits of the sample in order to evaluate an excess or a deficit.

EVALUATION Of MAXIMUM KYPHOSIS AND MAXIMUM LORDOSIS

The sacrum basis is the lower limit of lordosis. The slope of the sacrum basis on the horizontal varies from 19° to 65°, the average of the sample being 41° and the standard deviation 35°.

Between the maximum lordotic curve and the maxi­mum kyphotic curve there is an intermediate verte­bral body whieh is tilted most from the horizontal in the thoracolumbar region (Figure 2) This intermediate vertebral body ean be found at five different levels, three of which are more common: LI in 1/3 of cases, Tl2 in 1/5 of cases, and L2 in 1/5 of cases.

For each individuaJ, there exists a màximum ky­phot,ic curve, which in our study is measured accord­ing to the ang,le between the upper border of T4 at th top and the lower border of the intermediate vertebral body at the bottom. ln the same way, there exists a maximum lordotie curve, mcasured between the upper border of the interrnediatc vertebral body at the top and the sacrum base at the bottom.

These maximum kyphosis and lordosis are the main object of this preliminary study. For maximum

Fig 2. Oeterminlng sacrum base. maximum lordosis. i lermedlale vertebral body. and maximum ky­phosis. L1 = 33%; T12 = 22%; L2 = 21%.

S

L5

S

0 (0) 0

-35(-21)-10 fi

L5

--27(-81 5 b

L4

Table

L3

1. Reciprocal Anglulation of Each Vertebral Body'

L2 LI T/2 Til TIO T9 TB Tl T6 T5 T4

L4 -53(3~)-18 -2(21)-10 -10(-V 8

L3 -64(-47)-26 -54(-33)-16 -26(-12)-0a e s

-7(1110

L2 -74(-51)-28 -62(-4~H9 -36(1~)-6 -17(-~) 7 -5(1)11

L1 -79(-56)-33 -69(-42)-18 -44(-21)-2 \0 10 9

T12 -82(-55)-32 -72(-41)-19 -43(-20) 1 11 ,. 10

-23(--8)11 7

-27(-7)13a

-13(0)13 5

-16(1)20T

-3(5)12 3

-7(5)175

-5(5)H il

T11 -83(-53)-28 -70(-39)-17 -46( 18) 3 '1 II 10

T10 -81(-51)-24 -68(-37)-14 -43(-16) 9 '1 '2 11

-30{-5)15 G

-28(-3)21 JO

-21(3)23 a.

-22(5)29 9

-11(7)20 e

-14(9)25 7

-7(7)17::,

-12(9)22 ô

-3(5)12:\

-5(6)17 ~

-5(3)11 .1

T9 -80(-48)-23 -70(-35)-10 -42(-13)11 11 12 11

-29(-0)23'c

-23(8)33 -15(12)30 -12(11)29 lD .& 1

-5(9)27 6

-6(6)24 :>

-4(4)22 :.:

T8 -81(4,1)-20 -71(-3,1)-6 -36(-,~)14 -28(4)26 -22(1,~)40 -14(1~)37 -11(1~)36 -4(liJ36 -6(lS)37 -5(~)36 -9(4)14

Tl -68(-39)-13 -58(-25)-2 -32(-4)19 -23(9)27 -17(17)37 -9(21)44 -6(21)44 1(18)47 1(15)44 4(13)350(-10)22 -6(5)20 j 1 ~2 11 '0 ~o e :e 7 Î' 6 4 Il

T6 -62(-3,~)-8 -12(-2,Q) 2 -27(,~)27 -18(\~)35 -12(2~)45 -4(2:)51 -2(2~)52 5(21)55 5(2~)51 6\1~)44 4(1 ~)29 1(1 ;')22 -6(4J 14

T5 -61(-28)-3 -51(-14)10 -29(7)32 -21(20)41 -15(28)51 -6(32)56 -4(32)56 3(29)60 4(26)56 7(24)52 6(21)36 6(16)31 -2(10)23 -6(4)19\2" 12 \ 1 \' 11 '0 9 ~ Cl 8 ë :.> ,i 4

T4 -51(-23) 4 -38(-10)17 -20(12)36 -5(25)47 1(33)57 9(37)61 12(36)60 12(34)63 7(31)61 7(29)60 8(25)42 6(21)38 1(14)26 -2(9)17 -8(4)11 '2. 12 :1 1~ 'l 'c W 9 9 9 7 6 ~ r. 4

. Table of references based on observalion of 100adults, aged 20-29 years (43 women, 57 men) Values to the left are min:mums; values in parentheses are the means' values to the rlght are maximUms; values below are standard devlatlons Positive values are kyphosls; negatlve values are lordosis.

JJ

~'I =ij ,1

JJ 0 (') ~ r ~ Z C> c: r ~ -l <ë a

rZ C

II~ Cf)

'"---i :I> zGJ cz :;:::I> IJJ m:Il , :IJ:I>

"m ---i

<ô:I> 'ex> r '"

~ , lA; -.....J

SPI NE • VOLUME 7 • NUMBER 4 • 1982

338 RECIPROCAL ANGULATION • STAGNARA ET AL

~o

29 :0

29 205. 2'5 24

~i 2\ 2/}. 13.

1 18. 17. 1~.

17 1715 1614.

13 12 Il 10 ~. 8. 7 6 8 5. 4. J 2 1.

A 18 U"1:i 11"6 ,...... \; ....

l'S.

\4. , 15

13.

12

Il.

18.

9.

98.

7b

'5

4.

43 4 Z.

1.

B 18

1':1

14

13. 14

12. 13 1

Il

18.

9. 10

8.

7.

6. 7 1

'5

4

J 4

2. 3 3 1.

e 66

'Fig 3. Sacral base slopes. A. 100 French people; B. 43 French women; C. 57 French men.

lordotic curves, the values are between - 32. and - 84°; the average is - 50°, and the standard deviation is 30°. For the 43 women and 57 men making up the French sample, the average was not different accord­ing to sex. For maximum kyphotic curves, the values

3ll 29. ZOo ~7.

26. 25. 24. ?3 2522. 21. N. 1'. 18.

1917. 16. l'. 14. 13. 12. 113 1411 18. 11,.

108. 7. 6. 5. 4. 3. 2. 1.

A

15.

14

13. 14 12.

11.

18.

9.

8.

7 8 7

5

6.

6 4.

3

32.

2

B -32 -84

15. 1

14

13.

12.

Il

18 11 11 11

3.

8.

7 8

76.

5.

4.

3. 4 1

2.

1.

e

Fig 4. Maximum lordosis. A. 100 French people: B. 43 French women; C. 57 French men

of the angles between T4 and the intermediate verte­bral body can vary between + 7° and + 63°, with an average of + 37", and are the same for both men and women.

The standard deviations are considerable, and the

••

--

39

3

SPINE • VOLUME 7 • NUMBER 4 • 1982

'RECIPROCAL ANGULATION • STAGNARA ET AL 339

zao 1-0­2321. 28. 221 1'. 1•. 17. 1'. ­115. 1614. lJ. ,...- ­12. 1311. 1•. ,...- ­

••,.

10 7. ....--,6. 7, S. Ir ­4. 5J. 2. 1. . . . . . .. m ... ,..,. . . . .

za. lM. 23. .-- ­

18 2) 5 29 }4 S 40 45 5 51A 7 '''' 62

15.

14.

. 13.

12.

Il.

18.

109.

8. 9 87.

76.

65.

4.

2.

21.

,..,7 .,-, ,.. ,.',~".B 82

12. 1313 u. 1•. ,.

~

87.

6. i- ­6S.

4. 5 1-­-4 1 43.

2.

i. 1Il .'11 • . .JJ. ... . , .

15.

14.

13.

,.. ,..,e 7 62

Fig 5. Maximum kyphosis. A. 100 French people; B. 43 French women; C. 57 French men.

subjects furthest from the average va'iue are far from being handicapped. For example, a male subject with a maximum kyphosis of 14° and a vital capacity of 6.6 1 is an athlete with good performances in a number of djjferent fields, and, at the other end of the scale, a

20 -

" 16

12

10

A

15

14.

13 .

12.

Il.

Ill. 11 9.

a. 9

6. 7

5 6

4.

4i 3~.

1.

B

I~

14

13

20 ­18 18 18

-9

f----,

~ 6 5

1 2 1 2 !Tl

12. 13 Il 12 le.

9.

a. 9 7.

6 7

'5

4 5 4

2. 3 1. 1 2

,

e - 32 - "., -5

Fig 6. Most tilted vertebral body A. 100 French people. B. 43 French women; C. 57 French men.

male subject, with a maximum kyphosis of 63° and a vital capacity of 5.2 1 is an excellent mountaineer. Faced with such great variations, the study of the distribution of individual deviations became impera­tive in order to see whether they could be plotted on a

••

• • • •

•

• • ••

40

SPINE • VOLUME 7 • NUMBER 4 • 1982

340 RECIPROCAL ANGULATION • STAGNARA ET AL

• 77

• 71

• 65

62

5'

56 • 5' • 50

., • • • ,. ., ,. ')7• 2 5127 '0 '6 ••"

Fig 7. Correlation between sacral base siopes (horizontal axis) and maximum kyphosis (vertical axis) (43 French women).

Gaussian curve or whether they were irregular. The values of the sacrum slopes, the maximum lordotic and the kyphotic curves have been distributed in ten groups; each group brings together a certain number of cases of the sampie represented by the histogram columns.

The sacrum base slopes between 18° and 66° have been spread over ten groups of 4.S0 (average 41°) (Figure 3). For the 100 French men and women together, the distribution is relatively symmetric, but, if we divide the sexes, we observe for the 57 male subjects a deficit arounrl 45°, whereas many of the 43 females subjects are close to 40°. The maximum lor­dotic curves between - 32° and - 84° have been distributed in ten groups of 5.2° (average 50°) (Figure 4). As far as the French subjects are concerned, the distribution is fairly irregular. Therc are a good num­ber of cases between 45° and 70°. If we divide the sexes, the histograms are quite different. Female sub­jects are grouped around 50° to 60°, whereas male subjects form a compact block up to 75°.

The ltalian check group shows one particlllarity: for 19 female sllbjects in the column from 58° to 63°, right in the middle of the histogram, there are no cases. The sample, however, is not sufficiently large for us to attribute any special significance to this fact other than its status as an illustration of the extreme variability of individual cases.

The maximum kyphotic curves between + 7° and + 62° have been divided into ten groups of 5,50 (average 37°) (Figure 5). As far as the 100 French subjects are concerned, there is a c1ear grouping of the values between 30° and 50°. Male and female subjects show practically the same distribution. The ltalian check groups are less regular. In general, the ltalian subjects have slightly more accentuated curves than the French, and this could be a Medit rranean charac­teristic, but a larger sample is necessary before any conclu ions can be drawn.

Between kyphosis and lordosis, the intermediate

vertebral body, the most tiIted of the thoracolumbar region, shows a slope from the hor,jzontaJ, varying according to the individual, from - 32° to - 5°. (backward sloping), with an average value of - ISO. The distribution in ten groups of I.So shows a greater irregularity in female subjects (Figure 6).

CORRELATIONS BETWEEN KYPHOSIS­LORDOSIS AND SACRAL BASE SLOPES

ln 1951, A. Delmas, a French anatomist, described three types of spinal morphology: subjects with marked curves, subjeets with slight curves, and inter­mediate types. 1 With the "living" documents we brought together, it was possible to study the correla­tions of our three measurements (sacrum base, maxi­mum kyphosis and maximum lordosis) for each indi­vidual case. The morphologie types appear much more varied than those described by Delmas.

The computer presents us with a "cloud of points" and allows ellipses covering the most characteristic groupings to be descri'bed. Given the differences in distribution between male and female subjects in the histograms studied, we have treated each sex sepa­rately. In order not to overload the present paper, we give only the documents concerning the 43 French women.

CORRELATIONS BETWEEN SACRAL BASE SLOPE AND MAXIMUM LORDOSIS

As one might have expected, this correlation is quite close and can be descnibed in a long oblique ellipse. There is a marked correlation for the male sllbjects, a little less close for the female subjects (Figure 7).

CORRELATIONS BETWEEN MAXIMUM LORDOSIS AND MAXIMUM KYPHOSIS

One couId expect a close correlation, since a consid­erable lordotic curve seems to cali for a compensatory kyphotic curve as a corollary. Our documents show

" 50 1

.7

.1

'B

" • " 29 . ~

•

26

" 44 47 50 50J '56 59 62

Fig 8. Correlation between maximum lordosis (horizontal axis) and maximum kyphosis (43 French women).

SPINE • VOLUME 7 • NUMBER 4 • 1982

RECIPROCAL ANGULATION • STAGNARA ET AL 341

" * ,u

* * " **44 * 41

JB

* " * Jl

* * 29

26

*

**

*é*5Y*

2.\

27 JO JJ J6 J9 42

Fig 9. Correlation between sacral base slopes and maximum kyphosis (43 French women).

widely disposed "clouds" for men and women, but in the latter group the search for characteristic grouping ellipses displayed two zones separated by a thinly inhabited area. This could suggest that, whatever the maximum lordosis, the maximum kyphosis can equal­Iy weil be le s marked, or more marked relative to an average valu which we again observe as having little significance (Figure 8).

CORRELATIONS BETWEEN MAXIMUM KYPHOSIS AND SACRAL BASE SLOPE

Dispersion is even greater here. However, two zones of c ncentration are found. Rounder ellipses show considerable dispersion even within these zones (Figure 9).

Numerous other measurernents are being studied, including morphology of the sacrum and height of dises and vertebral bodies. This will make the study of

ther correlations possible.

DISCUSSION

To the best of our knowledge, no table of reference for the reciprocal angulation of vertebral bodies in standing humans existed prior to this study. What we propose is a protocol for the X-ray examination of the subject in an ea ily reproducible standing position. In roentgenograms taken without these precautions, vari­ations can occur, above ail, in the lumbar area.

Our sample is a modest ne. ur observations shouJd be checked by extending the study under the same conditions. The geographic origin of the sample should be m ntioned in e it se ms tbat this may be a factor of variation in pinal morphology.

There is very Httte differenc betwe n the sexes, the average values being sup rimposable. The lordotie appearance of feminine subjccts, no doubt is due to a more marked curve of the buttocks than we usually find in maIe subjects.

The span of a po sible value of maximum kyphosis

and lordosis in subjects with no spinal disease is considerable. The distribution of these values is often irregular within the extreme limits of the sample. Il is therefore unreasonable to speak of normal kyphotic or lordotic curves. The average values are only indicative and not normative. The cxtreme values show the morphologic latitude of the human species.

For smaLler segments, the deviat>ions are far less marked. Consequently, for angular kyphosis and lor­dosis, the reciprocal angk table can be used as a reference.

Each subject has a unique spinal "physiognomy," just as he has a unique facial "physiognomy."

A study of larger samples should allow more compli­cated morphologie types than those proposed by Del­mas to be distinguished. The functional values and patho.logic predispositions of these different types re­main to be studied. Where attenuated or accentllated kyphosis or lordosis are concerned, one should avoid using such adjectives as "excessive," "insufficient," or "pathologic" and, even more so, "abnormal."

Greek mythology offers us a sign,ificant legend. Procustus, a bandit, after having robbed his victims laid them on his bed (" Procustus' bed" is a proverbial saying), considered as being of standard (mean), or normal length. Those who were too short were stretched to the "right" lenglh, and those who were too tall were cul down lo size (Figure 10). In orlhopae­dics, let us avoid the Procuslus complex, lel us nol be

Fig 10. Procustus' complex. A (above). Procustus' bed. B (below). "Normal" spine.

42

Sf'INE • VULUME 7 • NUMB~R 4 • 1982

342 RECIPROCAL ANGULATION • STAGNARA ET AL

"mean"-minded (in ail senses of the word) without discernment.

SUMMARY In order to establish a sagittal plane curve reference

table for standing subjects examined laterally, we determined an easily reproducible standard posture. A sampie of 100 healthy subjects from 20 to 29 years of age, was chosen (43 women, 57 men). The reciprocal angulations of each vertebral body in relation to the others were fed into a digitalizer and studied by computer. The study particularly concems maximum kyphosis, maximum lordosis, sacral base slopes, and the tilt of intermediate vertebral bodies. The disper­sion of the results is remarkably wide and, within the extreme values, the distribution is irregular. Individual correlations of these values are often dispersed, but certain groupings can appear, allowing an approach to spinal morphotypology. For considerable lengths, av­erage values cannot be used as norms, given the wide

span of values. Only the extreme limits are useful for the appreciation of curves as excessive, insufficient, or inverted.

REFERENCES 1. De1mas A: Types rachidiens de statique corporelle.

Revue de Morphophysiologie humaine, 1951 2. Dran G: Etude radiologique et clinique des cour­

bures sagittales des segments thoracique et lombo­sacré de la colonne vertebrale. Thèse, Lyon, France, 1979

Address reprint requests to P. Stragnara

Centre De Readaptation Fonctionnelle Des Massues 92 Rue E. Locard Lyon .5

69322 Lyon Cedex 1 France

Accepted for publication March 10, 1981.