SPINAL CURVES AND HEALTH A SYSTEMATIC CRITICAL

LITERATURE REVIEW
SPINAL CURVES AND HEALTH: A SYSTEMATIC CRITICAL

REVIEW OF THE EPIDEMIOLOGICAL LITERATURE DEALING
WITH ASSOCIATIONS BETWEEN SAGITTAL
SPINAL CURVES AND HEALTH
Sanne Toftgaard Christensen, MD, DC,a and Jan Hartvigsen, DC, PhD b,c
ABSTRACT
Objectives: The purposes of this study were to (1) determine whether sagittal spinal curves are associated with health in
epidemiological studies, (2) estimate the strength of such associations, and (3) consider whether these relations are likely
to be causal.
Methods: A systematic critical literature review of epidemiological (cross-sectional, case-control, cohort) studies
published before 2008 including studies identified in the CINAHL, EMBASE, Mantis, and Medline databases was
performed using a structured checklist and a quality assessment. Level of evidence analysis was performed as outlined
by van Tulder et al (Spine. 2003;28:1290-9), and the strength of associations were determined using the procedure
outlined by Hemingway and Marmot (BMJ. 1999;318:1460-7). Quality of the included articles were assessed by our
own scoring system based on the STrengthening the Reporting of OBservational studies in Epidemiology checklist.
Studies scoring maximum points (4/4 or 3/3) were considered to be of higher quality.
Results: Fifty-four original studies were included. We found no strong evidence for any association between sagittal
spinal curves and any health outcomes including spinal pain. The included studies were generally of low methodological
quality. There is moderate evidence for association between sagittal spinal curves and 4 health outcomes as follows:
temporomandibular disorders (no odds ratios [ORs] provided), pelvic organ prolapse (OR, 3.18; 95% confidence interval
[CI], 1.46-96.93), daily function (OR range, 1.8-3.7; 95% CI range, 1.1-6.3), and death (OR, 1.40; 95% CI, 1.08-1.91).
These associations are however unlikely to be causal.
Conclusions: Evidence from epidemiological studies does not support an association between sagittal spinal curves and
health including spinal pain. Further research of better methodological quality may affect this conclusion, and causal
effects cannot be determined in a systematic review. (J Manipulative Physiol Ther 2008;31:690-714)
Key Indexing Terms: Review, Systematic; Spinal Curvatures; Lordosis; Kyphosis; Back Pain; Neck Pain; Health;
Chiropractic
ome health care providers believe that sagittal spinal
a
Medical Intern, Clinical Locomotion Science, Institute of
Sports Science and Clinical Biomechanics, University of Southern
S curves have an influence on health for example in
relation to back pain and neck pain.1,2 At the same time,
others believe that the relationship between sagittal spinal
Denmark, Odense M, Denmark. curves and health is not of great importance.3-5 Consequently,
b
Professor, Clinical Locomotion Science, Institute of Sports patients may receive contradictory information, and students
Science and Clinical Biomechanics, University of Southern Den- in different health care fields may be taught dissimilar
mark, Odense M, Denmark. information. In a world where the basis for health care relies
c
Nordic Institute of Chiropractic and Clinical Biomechanics Part heavily on published evidence, such divergence in opinions
of Clinical Locomotion Science, Odense M, Denmark.
Submit requests for reprints to: Jan Hartvigsen, DC, PhD, is unfortunate although probably common in many fields.
Professor, Institute of Sports Science and Clinical Biomechanics, Authors of nonsystematic reviews dealing with this
Campusvej 55, 5230 Odense M, Denmark subject also arrive at conflicting conclusions. Harrison
(e-mail: [email protected]). et al2 conclude that an ideal normal spinal model does
Paper submitted August 3, 2007; in revised form July 7, 2008; exists and that “normalization” of the curves may result in
accepted July 20, 2008.
0161-4754/$34.00 reversal of, for instance, back pain and sciatica,6 whereas
Copyright © 2008 by National University of Health Sciences. Haas et al7 questions this assumption. Troyanovich et al1
doi:10.1016/j.jmpt.2008.10.004 argue that treatment aiming at obtaining ideal sagittal spinal
690
Journal of Manipulative and Physiological Therapeutics Christensen and Hartvigsen 691
Volume 31, Number 9 Spinal Curves and Health
curves is indicated even beyond symptom relief, whereas on $ or risk or pain or instab$ or angle$ or model$ or
the other hand, Balagué et al3 find that there is no significant linear or mechanic$ or translation$ or mathematic$).
association between sagittal postural faults and low back mp.N” (n = 3620).
pain in children and adolescents. Gay4 states that the In Medline, search rubrics were:
determining factors of cervical curvature are incompletely (1) “(kyphos* or lordos*) and (model* or angle* or linear
understood and claims that more studies are needed to further or mathematic* or translation* or mechanic*)” (n =
characterize the influence of posture and trauma on the 1585).
cervical curve and pain. Ames5 finds the literature contra- (2) “[b(curvature* or align* or posture*) and (vertebra*
dicting with regard to posture in relation to low back pain or spin* or explode “spinal-diseases” or explode
and also recommends more research. “spine”)N or b(explode “spinal-curvatures”)N] and
When confusion exists in an area of general interest, a [explode “epidemiologic-study-characteristics”] (n =
systematic literature review is warranted. We therefore 3650).
decided to perform such a systematic review of epidemio- (3) “Spin* adj curvature*” (n = 426).
logical studies dealing with the relationship between sagittal (4) “Chiropractic biophysic*” (n = 17).
spinal curves and health in humans. The purposes of this
review are to (1) determine whether sagittal spinal posture is Titles and abstracts of the identified articles (n = 11 949)
associated with health in epidemiological studies, (2) were carefully screened by the first author for possible
estimate the strength of these associations, and (3) consider inclusion in the study. Potentially relevant articles (several
whether such relations are likely to be causal. hundred) were retrieved and read, and a final decision for
inclusion was made by the 2 authors together. Also reference
lists of included articles and of review articles were carefully
METHODS screened for possible additional studies.
We included cross-sectional, case-control, and cohort
studies comparing the sagittal curves of the human spine at all Abstraction of Information
ages to any health phenomenon. Studies dealing specifically A checklist was produced and used for abstraction of core
with scoliosis were excluded. Only full reports written in information from each article. Items included source, study
English were included. No letters or abstracts were included. design, characteristics of participants, measurement method
of sagittal curves and its validation, health parameter
investigated, statistical analysis, and conclusions. The 2
Search Strategies
authors independently completed one checklist per article.
Relevant articles published before January 1, 2008, were
The completed checklists were subsequently compared, and
identified by computerized searches in the databases
discrepancies were resolved by discussion until consensus
CINAHL (1982-2007), EMBASE (1974-2007), Mantis
was reached.
(1966-2007), and Medline (1966-2007). The search strate-
gies were developed in collaboration with a health science
research librarian. Assessment of Quality
Recently, there has been a focus on the reporting of
In CINAHL, search rubrics were as follows: observational studies via the STrengthening the Reporting
(1) “(kyphos* or lordos*) and (model* or angle* or of OBservational studies in Epidemiology initiative, and
linear or mathematic* or translation* or mechanic*)” this year's guidelines were published.8 These guidelines
(n = 487). are intended to improve reporting to better assess the
(2) “[b(curvature* or align* or posture*) and (vertebra* strengths and weaknesses of individual studies and not as a
or spin* or explode “spinal-diseases” or explode quality assessment tool per se. On the basis of these
“spine”)N or bexplode “spinal-curvatures”] and recommendations, we pilot-read a number of the included
[explode “epidemiology”]” (n = 269). articles, and we soon realized that the quality of reporting
In Embase, search rubrics were as follows: among the studies included in this review was very low.
(1) “(kyphos* or lordos*) and (model* or angle* or We therefore developed a simple 4-point quality score
linear or mathematic* or translation* or mechanic*)” evaluating basic features of the included studies based on
(n = 1513). the domains in the STrengthening the Reporting of
(2) “b(curvature* or align* or posture*) and (vertebra* or OBservational studies in Epidemiology statement best
spin* or explode “spine-disease” or explode “spine”)N pertaining to our research questions. After having
and bexplode “epidemiology”N” (n = 382). appraised 10 articles, we evaluated and found that (1)
In Mantis, search rubrics were as follows: our scores were identical and (2) the simple system did in
(1) “b(posture$ or curvature$ or kyphos$ or lordos$). fact result in a dispersion of scores and thus distinction
mp.N and b(disease$ or epidem$ or health or patholog between articles.
692 Christensen and Hartvigsen Journal of Manipulative and Physiological Therapeutics
Spinal Curves and Health November/December 2008
The 4 criteria and operational definitions were as follows: • Moderate evidence: provided by generally consistent
findings in one high-quality study and one or more low-
(1) Participants. quality studies, or in multiple (N1) low-quality studies.
• For cohort and cross-sectional studies, description • Insufficient evidence: only one study available or
of the study population including selection and inconsistent findings in multiple studies.
participation/response rate.
• For case-control studies, description of cases and of This method has previously been used in the assessment
selection of controls; a minimum of one age and of level of evidence in clinical and epidemiological studies in
sex-matched control per case. relation to low back pain.10,11
One point was given if the aforementioned criteria We then assessed the strength of the associations for all
were met. health outcomes according to the method used by Hemi-
In case of register-based studies (eg, hospital files or ngway and Marmot12 as follows:
public registries), no point was given. Accordingly, such
studies could achieve a maximum of 3 points. • No statistically significant positive association: P N .05
(2) Exposure. Any validation of the method used for the or OR or RR of less than 1 or 95% confidence interval
sagittal curve measurement reported. Any type of (CI) below or straddling 1.00, that is, OR or RR not
validation, i.e., anything relating to reproducibility or significantly different from unity.
validity) was accepted and awarded one point. One point • Moderate association: OR greater than 1.0 or RR of
was also given if a reference to another study dealing 2.00 or less.
with validation of the involved procedure was provided. • Strong association: OR or RR or more than 2.00.
(3) Outcome. Validation of the method used to measure
health parameter. Self-reported pain elicited one point. The strength of association for health outcomes showing
Studies dealing with any other health parameters strong or moderate evidence for an association to the sagittal
received one point for validation if a reference to a spinal curves was identified.
study dealing with validation of the measure was Assessment of Possible Causal Connection Between Spinal Curves and
given, if validation of the measure was carried out Health. Evaluation of whether exposures and effects are
within the study or if a diagnosis was made by a causally connected is difficult and requires as a minimum
licensed health care professional, that is, medical prospective cohort studies.13 Therefore, all cohort studies
doctor, chiropractor, or physical therapist. were pulled and studies showing positive association
(4) Statistical reporting. Odds ratio (OR) or relative risk between sagittal spinal curves and any health outcome
(RR) provided. One point was awarded if the magnitude were evaluated in relation to quality score and health
of the association was provided using OR or RR. outcome. Subsequently, results indicating a strong or
moderate evidence for an association between the sagittal
We did not perform any further evaluation of the spinal curves and health parameter were evaluated on an
measurement properties of these criteria as they are very individual basis for plausibility of a causal connection and
simple and easy to apply. the presence of adjustment for relevant confounders.
Analysis RESULTS
First descriptive items and results were extracted from the
Included Articles
checklists and tabulated along with the quality score.
Assessment of Quality. Studies scoring maximum points, i.e., Fifty-four including 20 709 participants were included in
3/3 or 4/4) were considered of higher (but not necessarily this systematic review. Thirty-six were identified through the
electronic searches and an additional 18 articles were
optimal) quality. Studies scoring less than maximum were
included after screening of reference lists.
considered to be of lower quality.
Assessment of Level of Evidence and Assessment of Strength of
Association. We are aware that level of evidence (LOE) criteria Abstracted Information and Quality Score
could influence results of systematic reviews.9 We therefore Table 1 gives a detailed outline of the descriptive items, the
first determined LOE based on the number, quality, and reported study results, and the quality scores. In summary, 20
outcomes of the studies as follows and subsequently cross-sectional,14-33 24 case-control,34-57 and 10 cohort
assessed the strength of the associations. Finally, we studies58-67 were included. A broad variety of procedures
combined the two. were used to measure the sagittal spinal curves. These were
grouped into 6 categories as follows: plain x-ray
• Strong evidence: provided by generally consistent films,17,18,21,25,29,32,33,35,37-41,43-47,49,51,53,55,57,59,63 magnetic
findings in multiple (N1) high-quality studies. resonance imaging,48 photographs,36,54,60 by eye,22,30,31,52,60
Table 1. Descriptive items, conclusions, and quality scores of 51 epidemiological articles dealing with associations between sagittal spinal curves and health
Volume 31, Number 9

Journal of Manipulative and Physiological Therapeutics
First author, Study Participants Participation Evaluation of Health symptom Statistics Conclusion OR Score
year, country design (cases/controls) sagittal curves and evaluation 1
(reference) and response rate 2
3
4
Brav, 1942, Case- Population: patients 62/35 X-ray: lateral LBP and sciatic NR Lumbar lordosis Not provided 0
USA35 control in an orthopedic/ projections of dorsal pain; no further was present twice as 0
radiology hospital and lumbal columna; information frequently in the
Cases: LBP and Response measure-method group of painful 1
sciatic pain rate: NR of lumbar lordosis: backs; decreased 0
unknown lordosis with
Controls: younger 1
flattening of this
than cases; no report
area of the spine
on matching
was 6 times more
often in the
backache cases than
in the controls
Anderson, Cohort Population: 423 X-ray: lateral chest Survival Correlations Degrees of kyphosis Not provided 0
1976, 70-89–y-old people projection; kyphotic time/death analysis are useful
Scotland59 referred by their Response rate: angle measured by Evaluated predictive criteria 0
general practitioner baseline, NR; intersection of through the in respect of survival
to Rutherglen consultative follow-up, 94% lines through anterior health time in elderly in
health center for older midpoint of T2/T3 departments' the elderly population
people and T11-T12 death-register
Selection: NR 1
0
1
Milne, 1983, Cohort Population: elderly 487 X-ray: lateral chest; Death 2-tailed The index of Not provided 0
Scotland63 (62-90 y) citizens standing; measure- t test kyphosis
of North Edinburgh method, index in men is related
Selection: random Response rate: of wedging Linear to mortality 1
selection from sample baseline, NR; Surveyor's regression 1
of 27 000 citizens follow-up, 54% flexicurve: 0
C7-lumbosacral
2
junction
Frymoyer, Cross- Population: all male 321 X-ray: lateral Current and χ2 test Measured lordotic Not provided 0
1984, sectional patients age 18-55 in projection previous LBP angle L1-L5 was 1
USA21 a family practice of of lumbar spine; diagnosed by similar for all 3 1
8000 people were sent standing Likert-type scale groups
a questionnaire about from McGill Pain
pain; 1221 (67%) Questionnaire
Christensen and Hartvigsen

Spinal Curves and Health
responded
Selection: random Response Measure-method: Spearman's 0
selection from the rate: NR Cobb-like method; correlation
2
responding population lumbar lordosis coefficient-
that was divided L1-L5, upper lumbar ANOVA
in 3 groups according lordosis L1-L3, techniques
(continued on next page)
693
694
Table 1. (continued)

3
4
to pain—moderate, lower lumbar with multiple
severe, or no pain lordosis L3-L5 comparisons
Hansson, Case- Population: heavy (200+200)/ X-rays: lateral Acute and χ2 test The distribution and Not provided 1
1984, control labor workers or 200 projection of chronic LBP range of lordosis 1
Sweden/ applicants for these lumbosacral region; does not vary 1
USA39 positions, who had supine position; in the patients
preemployment x-rays measure-method, with acute or
taken at an occupational goniometer; lumbar chronic LBP more
medical clinic in 1982 lordosis, L1(sup)- than in men
Cases: 200 with first Response S1(sup) without back pain 0
time LBP at time of rate: NR of the same age
3
x-ray examination,
200 with LBP causing
absence from work for
N6 mo, or people who
had x-rays taken as
part of disability
evaluation
Controls: 200 without
previous LBP or
disability; age and
sex-matched
Salminen, Cross- Population: Finnish 380 By eye: Matthias test; Neck and Logistic Subjects with Not provided 1
1984, sectional school pupils age 11, evaluation of thoracic back pain linear model hollow back 1
Finland31 13, 15, and 17 and lumbar spine experienced
Selection: random Response significantly more 1
selection from all rate: 97.4% often from 0
senior secondary symptoms than
3
school pupils those who had

no hollow back
Biering- Cross- Population: 60-y-old 666 X-ray: lateral projection LBP-interview Fisher exact There was no Not provided 0
Sørensen, sectional inhabitant from of thoracic and lumbar positive correlation 0
1985, Glostrup participating Response spine; supine, 88%; χ test
2
between LBP with 1
Denmark17 in a cohort study rate: NR standing, 12%; measure- or without leg pain
started in 1964 and method: not reported and thoracic
39 other participants kyphosis or lumbar
without further columna recta
November/December 2008
information
Selection: partially Wilcoxon- 0
reported Mann-
Whitney
Kruskal- 1
Wallis test
Dieck, Cohort Population: women 1948 Lateral LBP, midback χ2 test Increased degree of Not provided 0
1985, who were members Response rate: photographs pain, or neck pain T test kyphosis or lordosis 0
Volume 31, Number 9
USA60 of graduating class baseline, NR; of posture were evaluated by Correlation did not increase 1
1957, 1958, and 1959 follow-up: questionnaire coefficient the risk of
of Eastern US women 52.7% 25 y past baseline subsequent
college; no back/neck spinal pain
pain before age 18
Selection: all members 0
of the 3 graduating classes
1
During, 1985, Case- Population: patient (20+24)/52 X-ray: lateral LBP-referred Student t test There was no Not provided 0
The control were referred to an Response projection because of LBP positive correlation 0
Netherlands37 outpatient clinic rate: NR of lumbar between patients
(no specified spine and with disk space
clinic type) for LBP pelvis; standing; narrowing or
Cases: 20 with space measure- without x-ray 1
narrowing of L5/S1, method: aberrations and
24 with no instantaneous the normal group
x-ray changes shape of
Controls: no the lordotic 0
matching reported curvature
1
Hertzberg, Cohort Population: 4 annual 302 During clinical Head, neck, and 2 × 2 table There was no RR = 1.5 0
1985, cohorts with a total Response rate: examination in back pain— χ2 test significant increased (CI = not 0
Norway61 of 316 persons who baseline, NR; 1970-73 by questionnaire risk for cervical reported); RR 0
at the age of 16 follow-up: district physician, and lumbar for cervical 1
completed 9 y of 96% a posture pain at follow-up pain in
school education assessment was for persons relation to
in Norway 1970-73; made—no with increased posture
302 were included description of thoracic kyphosis
in the follow-up the procedure at baseline
in 1982
RR = 1.0 1
(CI = not
reported); RR
for lumbar pain
in relation
to posture
Selection: NR
Pope, 1985, Cross- Population: 321 X-ray: lateral Moderate and χ2 test The 3 groups Not provided 0
USA29 sectional All male patients Response projection; severe LBP Pearson's showed no 1
age 18-55 in a family rate: NR standing; measure- diagnosed by correlation significant
practice of 8000 people method: Farfan's Likert-type Means of differences in 1
were sent a questionnaire method; lumbar scale from ANOVA lumbar lordosis
about pain; 1221 (67%) lordosis: L1-L5, McGill Pain techniques
responded L1-L3, and L3-L5 Questionnaire
Selection: 321 0
were randomly 2

selected from

3 groups (no LBP,
moderate LBP, severe LBP)
of an initial
population of 1221
695
696

3
4
Chow, Cross- Population: healthy 47 Surveyor's flexicurve Calcium Bone Pearson Unclear Not provided 0
1987, sectional white female volunteers measures from Index measured product-
USA19 Selection: NR Response C7-L5/S1; the by neutron moment 1
rate: NR index of kyphosis activation analysis correlation 1
(I/K) is given from coefficient
the measurements Level of Student t test 0
physical fitness
2
measured by
VO2 max at a
threadmill
Strength
performance
measured by
bench press
Bergenudd, Cross- Population: residents of 575 Spinal pantograph was LBP diagnosed Regression Back pain was not Not provided 0
1989, sectional Malmö City used to evaluate by pain drawings statistics related to degree of 1
Sweden16 Selection: all third Response the sagittal curves ANOVA kyphosis or lordosis 1
graders in the elementary rate: 69% 0
schools in 2
Malmö in 1938
Bryner, Cross- Population: 124 124 X-ray: lateral LBP as chief χ2 test There is a weak Not provided N/A
1992, sectional radiographs from Response projection of lumbar complaint Pearson association 1
Australia18 files in 3 chiropractic rate: 100% spine; measure- correlation between the lumbar 1
teaching clinics of Phillip (register-based) method: Cobb-like coefficient curve angle and the
Institute of Technology method; lmbar presence of symptoms
Selection: random curve angle L1-L5 0
selection from the files
2/3

Griegel- Cross- Population: NR 88. Thoracic kyphosis Thoracic, cervical, Spearman Inconclusive Not provided 0
Morris, sectional Selection: Volunteering Response in standing position and shoulder pain rank-order 1
1992, subjects rate: NR was evaluated by eye correlation
USA22 χ2 test 1
0
2
Waddell, Case- Population of cases: 120/77 Inclinometer in LBP ANOVA Spinal posture Not provided 0
1992, control referrals from Response standing position. Impairment Unpaired was weakly and 1
Scotland56 National Health Service rate: NR Thoracic kyphosis: t test inconsistently related
to an orthopedic outpatient T2/T3-L1/L2. Lumbar to the presence or
clinic because of LBP lordosis: L1/L2-S2 absence of back
Population of controls: Discriminant pain; lumbar 1
patients with minor analysis lordosis was 0
hand injuries, friend/ not significantly
Volume 31, Number 9
family of patients, different in patients
or hospital staff and normal subjects
Cases: LBP N 3 mo; with although kyphosis 2
or without referred pain to was; spinal postures
buttocks or thigh of lordosis and
Controls: no matching kyphosis were not
reported; stratified by related to disability
age and sex in 5-y
age bands from 20-55;
no present LBP,
no history of LBP
N 1 mo
Nagasawa, Case- Population: NR 372/225 X-ray: lateral Tension-type Unpaired The results suggest Not provided 0
1993, control projection of headache t test that a straightened
Japan/ Cases: tension- Response cervical spine; according to Wilcoxon cervical spine may 1
Australia49 type headache rate: NR measure-method: International rank sum test play an important
Controls: Ishihara's method Headache Society Welch's test role in the 1
Sex and Classification pathogenesis 0
age-matched criteria of tension-type
headache and its
accessory symptoms 2
Nissinen, Cohort Population / selection: 1060 Spinal pantograph LBP Logistic There was no OR, 1.20 0
1994, grade 4 school children was used to measure regression association between (0.97-1.50) ♂
Finland64 in Helsinki in 1986 sagittal spinal profiles model thoracic kyphosis
Selection: all grade Response rate: in standing position χ2 test or lumbar lordosis OR, 1.09 1
4 children baseline, NR; and LBP (0.81-1.28) ♀
follow up, 81% 2-sided ORs for 1
t test association 1
between
kyphosis and
LBP
OR, 0.84 3
(0.61-1.16) ♂
OR, 1.08
(0.84-1.33) ♀
ORs for
association
between lumbar
lordosis and
LBP
Jackson, Case- Population: NR 100/100 X-ray: lateral LBP ≥ 6 wk; T test Total lordosis was Not provided. 0
1994, control Cases: LBP ≥ 6 wk Response projection visual analog Pearson significantly less
USA43 rate: NR of the entire spine; scale correlation in the low back

Controls: sex-matched standing; measure- ANOVA pain patients. 1

method: Cobb; Newman- Patients tended to 1
thoracic kyphosis Keuls stand with less 0
T1(sup)-T12(inf) multiple distal segmental
2
and lumbar lordosis comparison lordosis
L1(sup)-S1(sup) test
697
698

3
4
Christie, Case- Population: recruited 39/20 Photographs of Acute LBP 1-way In standing, patients Not provided 0
1995, control from selected medical posture with small (b6 mo and no ANOVA with chronic
Canada36 institutions and a Response balsa wood pointers LBP for the χ2 test pain exhibited 1
university campus rate: NR used as markers on past 12 mo) an increased
Cases: chronic group C7, T12, and L5; and chronic LBP A Tukey post lumbar lordosis 1
with LBP of continuous standing and seating; (continuous or hoc analysis compared with
or recurrent nature thoracic kyphosis recurrent nature T test controls; patients 0
for N6 mo; acute group and lumbar lordosis for N6 mo) with acute pain
with LBP were measured with had an increased
b 6 mo and no LBP method of Flint thoracic kyphosis
for the past 12 mo compared to
Controls: no LBP during Linear controls in 2
the past year and never discriminant standing and
LBP N 1 mo analysis sitting positions
Power analysis
Refshauge, Case- Population: NR 18/18 Markers placed on Pain in the Independent The size of the Not provided 1
1995, control Cases: presenting for Response the spinous process cervical or 2-tailed t test cervicothoracic curve 1
Australia54 treatment of pain in the rate: NR of C4, C7, and T4 trapezius region in the sagittal plane 1
cervical or trapezius by palpation; 2 lateral during relaxed
region but not for photographs are standing is not
headache taken with a associated with, and
Controls: age and Polaroid camera; therefore, is unlikely 0
sex-matched cervicorthoracic to contribute to, the
3
angle is measured onset of pain
Lind, Case- Population: 48/48 X-ray: lateral Uterine prolapse; Paired t test Thoracic kyphosis OR, 1.35 1
1996, control 412 consecutive Response rate: projection of thorax; clinical examination McNemar test appears to be associated (1.11-1.65) 1
USA47 patients who underwent 100% (register- measure-method: with uterine prolapse

either abdominal or based) Ferguson's method
vaginal hysterectomies
at North Shore University
Hospital in 1990-1993
Cases: uterine prolapse Conditional Thoracic kyphosis OR for 1
to or beyond the logistic regression was higher in association
level of introitus patients with uterine between each
Controls: no evidence Multivariate prolapse than additional 1
of uterine prolapse; conditional in controls degree of
4
matched for age, sex, logistic kyphosis
weight, menopausal status, regression and uterine
and hormonal status prolapse
Volume 31, Number 9
O'Brien, Cross- Population: volunteers 48 Kyphosis T1-T12 1 or more falls T test No significant Not provided. 0
1997, sectional ≥65 y recruited from Response and lordosis during the 1-y Spearman's ρ difference was 1
Canada28 family physician, rate: NR T12-S1 mesured before the study nonparametric test found between
geriatric, and home with an inclinometer fallers and nonfallers
care programs or senior in standing position on kyphosis
centers and lordosis
Selection: NR Balance: Mann Whitney 1
Berg balance U test
scale, functional Independent There was a weak 0
reach test, and t test but significant
2
timed get up and correlation between
go test upper thoracic slope
and all 3 balance
measures; increased
kyphosis gave
deteriorated performance
Ryan, Cross- Participants: 231 Kyphosis measurements: Interview: Spearman's Kyphosis, by both OR, 1.64 0
1997, sectional 231 community- qualitatively, by eye; self-report on correlations clinical and quantitative (1.03-2.61);
USA30 dwelling volunteers N quantitatively, distance health assessment, is OR for
age 59 recruited from Response from apex C7 to wall Test: 5-m walk Wilcoxon associated with association 1
a senior center in rate: NR and a timed t test diminished function, between
Baltimore city/county walk up and especially performance qualitative
Selection: NR down a flight χ2 test of mobility tasks measurement 1
of 14 stairs Multiple of kyphosis 1
linear and and heavy 3
logistic household
regression OR, 1.38
analysis (1.12-1.68);
OR for
association
between
quantitave
measurements
of kyphosis and
heavy household
Hein, Case- Population: NR 30/30 Sagittal thoracic Back pain N 7 wk LoA statistics The mean lumbar 0
1998, control Cases: back pain N Response and lumbar curves or N3 episodes curve depth was 1
England42 7 wk or N3 episodes rate: NR were measured by of back pain found to be Not provided
of back pain within a flexicurve with the within the last greater in
the last 6 mo method of Conner 6 mo symptomatic patients
Controls: age and and Breen compared with 1
sex-matched asymptomatic controls 0
No statistically 2
significant difference was

found according to the
thoracic curve
699
700

3
4
Jackson, Case- Population: NR 50/50. X-ray: lateral projection Symptomatic Student t test Patients with Not provided 0
1998, control Cases: 50 degenerative Response of entire spine; standing; degenerative ANOVA degenerative disorders 1
USA44 lumbar disk disease rate: NR measure-method: lumbar disk showed less total l
and LBP Cobb method; disease; ordosis than controls
Controls: no matching lumbar lordosis no further Pearson 0
reported L1(sup)-S1(sup) and information correlation
thoracic kyphosis coefficients
T1(sup)-T12(inf) Newman- 0
Keuls
1
multiple
comparison
test
Harrison, Case- Population: NR (50+50)/50 Digitized x-ray was Acute and F test Methods such as Not provided N/A
`1998, Cases: 50 cases with RR: 100% analyzed with GP-9; chronic LBP; Bartlett's test the 2-line posterior 1
USA40 control acute LBP (first time absolute rotation angle diagnosed by ANOVA tangent method, the 2- 1
LBP and b6 wk) and T12(inf)-S1(sup); medical history line Cobb method,
50 chronic LBP measure-method: and the elliptic ratio
(recurrent LBP or first Cobb method (2 line) parameter semiminor
time LBP N 6 wk) axis to semimajor axis,
Cases were matched to (Retrospective) b/a, can separate 0
controls in age, height, average pain subjects
weight, and sex from normal
Controls: 50 controls subjects; normal average 2/3
were recruited first, values or values from
and then the cases were the ideal lumbar model
matched to the controls could be used in spinal
modeling, in clinical
practice and as structural
outcomes of care

Adams, 1999, Cohort Population: volunteering 403 3Space Isotrak device: LBP, questionnaire Univariate Personal risk factors OR, 1.8 0
Switzerland58 health care workers Response lumbar lordosis (0, 6, 12, 18, 24, linear logistic (including reduced (1.0-3.2); 1
with no history of serious LBP rate: baseline, (skin marks on 30, 36 mo) regression lumbar lordosis) OR for
Selection: NR NR; follow-up L1-sacrum) measured explained up to 12% reduced 1
at 12 mo, 99%; at baseline of first-time LBP lordosis
30 mo, 90% Correlation and LBP
analysis
Linearity of 1
risk
3
Korovessis, Case-c Population: Greek 120/120 X-ray: lateral projection LBP N 2 mo Simple Lumbar lordosis Not provided 0
1999, ontrol nationality, no further of thoracic and regression was significantly greater
Greece45 information lumbar spine; standing; analysis in the controls; thoracic
Cases: LBP ≥ 2 mo Response measure-method: χ2 test with kyphosis was 1
rate: NR Cobb method; Yate's significantly greater in
thoracic kyphosis correction the LBP group
Volume 31, Number 9
Controls: age-matched T4(sup)-T12(inf) and Unpaired t test 1
lumbar lordosis 0
T12(inf)-S1(sup)
2
Tüzün, Case- Population: NR (50+50)/50 X-ray: lateral Acute (b6 mo) 1-way There was no statistically Not provided 0
1999, control projection of thoracic and chronic ANOVA significant difference
Turkey55 Cases: 50 cases with Response and lumbar spine; (N6 mo) LBP T test among the groups 1
chronic LBP N 6 mo; rate: NR standing; measure- χ2 test for angles of thoracic 1
50 cases with acute method: Cobb-like kyphosis or
LBP b 6 mo method; lumbar lumbar lordosis
Controls: no matching lordosis L1(sup/inf Pearson 0
[inconsistent product
description])- moment
S1(sup), thoracic Spearman 2
kyphosis T1(sup)- rank order
T12(inf) correlation
analysis
Mattox, Cross- Population: women 363 Posture measured Pelvic organ NR An abnormal change OR, 3.18 0
2000, sectional referred to a multicenter Response by Milne and Lauder's prolapse diagnosed in spinal curvature, (1.46-6.93); 1
USA24 for various complaints rate: NR technique with a by Pelvic Organ specifically, a loss OR for 1
of urinary incontinence semi-flexible rod Prolapse of lumbar lordosis, association
or pelvic organ prolapse from C7-L5/S1 Quantitation appears to be between
Selection: NR system a significant risk abnormal 1
factor in development spinal
3
of pelvic organ prolapse curvature
and pelvic
organ prolapse
Nguyen, Case- Population: women who 20/20 X-ray: lateral Uterovaginal Unpaired Women with advanced Not provided 1
2000, control were evaluated in the Response lumbosacral; standing; prolapse; 2-tailed uterovaginal prolapse 1
USA51 gynecology clinic at Harbor- rate: NR measure-method: clinical Student t test have less lumbar lordosis
UCLA Medical Center Cobb-like method; examination using 1
January 1997-June 1998 lordotic angle Baden and Walker Mann-
were asked to participate L1(sup)-L5(sup) classification Whitney
U test
Multivariable 0
logistic
Cases: Uterovaginal 3
regression
prolapse ≥ grade 2
Controls: Matched
according to age, number
of vaginal deliveries, BMI,
race, and menopausal status
Nicolakis, Case- Population of cases: 25/25 By eye: thoracic Craniomandibular Exact Significantly more Not provided 1

2000, control patients with craniomandibular and lumbar curves disorders Mann- postural abnormalities

Austria52 disorders (CMD), selected Response are evaluated as flat, Clinical diagnosis Whitney were noted in the 1
consecutively from subjects rate: NR normal, or increased U test patients when compared 1
consulting the CMD service kyphosis/lordosis to the control group 0
at the Department of Dentistry,
3
University of Vienna
Population of controls: NR The evaluation is
701
702

3
4
complex and counts
several different
parameters (Peterson
Kendall's principles)
Controls: age and
sex-matched
Nourbakhsh, Cross- Population/ 840 Flexible ruler: LBP N 6 wk or χ2 test There was no Not provided 0
2001, Iran26 sectional selection: Response lumbar lordosis 3 episodes of Logistic significant 1
subjects from 8 hospitals in rate: NR measured by LBP for 1 mo regression difference in
Tehran, Iran Youdas' during the last year analysis the degree
Selection: random selection method of lumbar 1
lordosis in 0
subjects with or
2
without LBP
Tsuji, 2001, Cross- Population: elderly 489 X-ray: lateral 1 episode of Unpaired There was a Not provided 0
Japan32 sectional Japanese (50-85 y) projection; standing; LBP within the t test significant difference
volunteers who visited Response measure-method: last 3 mo, χ2 test in lumbar lordosis 1
a health clinic rate: NR Cobb-like method; (acute LBP between the groups
Selection: NR lumbar lordosis: was excluded) Pearson's with and without LBP; 1
L1(sup)-S1(sup) and leg pain correlation lumbar lordosis was 0
radiating from test ca 4° less in LBP group
the low back
Visual analog 2
scale
Widhe, 2001, Cohort Population: every newborn 116 Debrunner's Age 15-16: T test Posture at age 5-6 or Not provided 0
Sweden66 baby in a period of 1 y at Response kyphometer at age LBP Linear 15-16 had no significant 1
a selected hospital rate: baseline, 5-6 and age questionnaire regression relationship to LBP

Selection: population was NR; follow- 15-16: thoracic χ2 test at age 15-16 1
divided in 2 by normal or up, 78% kyphosis T2/3-T11/12 ANOVA 0
minor foot deformities at and lumbar lordosis
2
birth; random selection T11/12-S1/2
from each group
Ng, 2002, Case- Population: NR 15/15 Inclinometer: lumbar LBP T test There was no difference Not provided 1
USA50 control Cases: back pain history Response lordose T12/L1-L5/S1; Visual analog in degree of lordosis 1
N 12 mo rate: NR standing scale between back pain 1
Controls: age, height, Roland-Morris patients and controls 0
obesity, physical activity, Disability
3
and sex-matched Questionnaire
Volume 31, Number 9
Korovessis, Case- Population: NR 100/100 X-ray: lateral LBP N 6 mo Unpaired Conclusion on Not provided 0
2002, control projection of lumbar t test correlations between
Greece46 Cases: LBP N 6 mo Response spine; standing; Simple linear LBP and sagittal curves 1
Controls: ethnic, work, rate: NR measure-method: regression is not provided 1
age, height, and Cobb-like method analysis and 0
sex-matched multiple
Lumbar lordosis 2
linear
T12(sup)-S1(sup)
regression
and distal lordosis
analysis with
L4(sup)-S1(sup)
post hoc
comparison and
Scheffé test
Correlation
coefficient
Nourbakhsh, Cross- Population: subjects 600 Standard flexible LBP N 6 wk χ2 test The size of Not provided 0
2002, Iran27 sectional were recruited from Response ruler: lumbar lordosis or 3 episodes Backward lumbar 1
5 hospitals in Tehran, rate: NR measured by Youdas' of LBP for 1 mo logistic lordosis 1
Iran; controls either method during the last regression is not 0
accompanied a patient year; no referred analysis associated
2
or were referred to leg pain with the
hospital for a occurrence
nonmusculoskeletal reason of LBP
Selection: NR
Rajnics, 2002, Case- 50/30 X-ray: lateral Disk herniation T test There was significantly Not provided 0
France/ control Population of cases: Response projection of the evaluated by Pearson less L1-L5 lordosis in 1
Hungary53 patients examined and rate: NR entire spine; controls MRI scan correlation cases compared with
treated in 4 French were standing, test controls
orthopedic centers position of cases
Population of controls: NR unknown; measure- Fischers z test 0
Cases: neurologic method: Cobb-like There was no 0
symptoms and evidence method significant differences
of disk herniation by MRI in the T4-T12 kyphosis
Controls: no matching Thoracic between the 2 groups 1
reported kyphosis
T4-T12 and
lumbar
lordosis
L1-L5
Murrie, 2003, Case- Population of cases: 27/29 MRI scan; supine Chronic LBP Mann- Angle of lordosis did Not provided 0
New control 27 patients referred to Response position with Whitney test not vary significantly 1
Zealand48 MRI clinic for chronic rate: NR supporting cushings between those with
LBP under knees and without LBP
Population of controls: 1

19 patients without LBP 0

referred to MRI clinic and
10 volunteering staff
Cases: chronic LBP Lumbar lordosis: 2
Controls: no matching L1/L2-L5/S1
reported
703
704

3
4
Evcik, 2004, Case- Population: NR 30/30 X-ray: cervical TMD N 6 mo Mann There was a decrease Not provided 0
Turkey38 control lordosis Whitney U test in the cervical lordosis
Cases: clinical findings of Response Measure-method: χ2 test in 16% of cases, 0
temporomandibular rate: NR not reported whereas in 2% 1
disorder (TMD), tender on of controls 0
palpation, asymmetrical
mandibular movements,
and joint sounds N6 mo
Controls: no signs of TMD; 1
no matching reported
Di Bari, 2004, Cross- Population: an entire 323 Occiput wall distance Dyspnea χ2 test Presence of kyphosis OR, 2.5 1
Italy20 sectional unselected home-dwelling, Response Difference between FVC (SS Multivariate was associated with a (1.1-5.8); 1
elderly (N65) population rate: 71.1% standing stature and KHS) risk higher prevalence of OR for
recorded in the city registry (SS) and knee dyspnea and lower association
office was enrolled height stature (KHS) FVC, FEV1, between
Selection: the whole FEV1 (SS Prevalence OR and FEV1/FVC kyphosis 1
population was included and KHS) and
FEV1/FVC. Logistic dyspnea 1
regression.
Multivariate OR, 3.3 4
regression (1.7-6.5);
OR
for association
between
kyphosis
and obstructive
ventilatory
pattern
OR, 2.3

(1.1-4.8); OR
for association
between
kyphosis and
restrictive
ventilatory
pattern
Norton, 2004, Cross- Population: patients in 227 Trigonometric Low back T test There was no difference Not provided 0
USA15 sectional 6 physical therapy clinics, Response method by Youdas pain ± leg pain in lumbar curvature 1
patients' friends and rate: NR et al: T12/L1-S2 between people 1
families, and volunteers with LBP and 0
through public people without LBP
advertisements
Selection: NR 2
Volume 31, Number 9
Kado, 2004, Cohort Population: participants 1353 Distance from Information Cox Older men and Relative 0
USA62 from the Rancho Bernardo Response occiput to table in of vital proportional women with hazard = 1.40 0
heart and chronic disease rate: baseline, supine position with status (death) hazards hyperkyphotic posture (CI = 1.08-1.81); 1
study established in 1972, 80%; follow-up, head in neutral position was collected analysis have higher relative hazard
which were alive in 80.5% annually mortality rates ratio for hyper
1988-1991 kyphotic posture
Selection: NR in relation to 1
death
2
Harrison, Case- Population: register-based (52+70)/72 X-ray measured with Cervical pain. 2-sample The cervical lordosis Not provided N/A.
2004, control subjects from a clinic in digitized sonic t test in both acute and
Canada41 Saugus, Mass Response system; measure- Variance chronic neck pain 1
rate: method: posterior analysis patients was found
Cases: acute neck pain 100% tangent and Cobb ANOVA to be hypolordotic 1
(b12 wk of pain and first (register- method; cervical Kruskal- 0
time of pain) and chronic based) lordosis C1-C7 Wallis
2/3
neck pain (N12 wk of pain and C2-C7
or N1 episode of neck pain)
Controls: participants from
a former study; no cranial,
neck, or brachial pain;
≤10-mm anterior
head translation;
no matching reported
Sinaki, 2005, Case- 12/13 X-ray: lateral Gait analysis Student t test Thoracic Not provided. 0
USA57 control Response projection of the Muscle strength Variance hyperkyphosis 1
rate: NR thoracic and lumbar analysis on a background
Population: community- spine; measure-method: of reduced
dwelling persons, Cobb method; muscle strength plays an
recruited on a volunteer thoracic kyphosis important role in
basis T2-T12 increasing body sway,
Cases: osteoporotic Computerized Pearson gait unsteadiness, and 1
persons with thoracic posturography correlation risk of falls in 0
hyperkyphosis test osteoporosis
Controls: persons 2
without osteoporosis or
hyperkyphosis; sex-matched
McAviney, Cross- Population: 277 x-ray 277. X-ray: full spine Cervical pain; χ2 test There was statistically OR, 18 (CI N/A
2005, sectional films from the Summer Response rate: series; measure- diagnosed from 2-sided significant association not reported); 1
USA25 Hill Chiropractic 100%. (register- method: posterior patients' record, Mann- between cervical OR for
Outpatients Clinic, based) tangent method; x-ray referral slip, Whitney pain and lordosis b20° association
Australia, from the past 7 y absolute rotation and radiology U test between
Selection: random angle C2-C7 report ROC curve cervical pain 1
analysis and lordosis 1

of 0° or less
OR, 2 3/3
(CI not
reported);
OR for
association
between
705
706

3
4
cervicogenic
symptoms and
absolute rotation
angel ≤20°
Kado, 2005, Cross- Population: surviving 1578 Distance from Functional status Logistic Older persons with Summation of 0
USA23 sectional participants aged ≥55 y occiput to table in diagnosed by regression. hyperkyphotic posture OR:
in 1988-1992 from a Response rate: supine position with questionnaire χ2 are more likely to have Range of OR, 0
cohort study established 80% head in neutral position physical functional 1.8-3.7
in 1972 of upper/middle difficulties
class residents
Selection: inclusion of the Grip strength Student t test Range of 1
whole cohort and chair stand CI = 1.1-6.3
ability Multivariable ORs for 1
diagnosed by model association
2
clinical tests between poor
physical function
and kyphosis
Poussa, 2005, Cohort Population: grade 1060 NR LBP Logistic There was no association Summation of 0
Finland65 4 school children in regression between thoracic OR:
Helsinki in 1986 analysis kyphosis or lumbar
Selection: inclusion of Response rate: lordosis and LBP Range of OR, 0
the whole cohort baseline, NR; 0.87-1.22
follow-up, 41% Range of 1
CI = 0.63-1.69
ORs for 1
association
2
between
kyphosis or

lordosis and LBP
Huang, 2006, Cohort Population: female 596 Distance from Fractures Logistic Hyperkyphotic posture Summation 0
USA67 participants from the occiput to table in diagnosed by regression may be an important risk of OR:
Rancho Bernardo heart Response rate: supine position with lat thoracolumbar analysis factor for future Range of OR, 0
and chronic disease baseline, NR; head in neutral position x-ray and fractures, independent 1.73-1.77
study established in 1972, follow-up, NR questionnaire of low BMD or fracture
which were alive in history
1988-1991
Selection: NR Range of 1
CI = 1.02-3.05
ORs for 1
association
2
between
hyperkyphotic
posture and
future fracture
Volume 31, Number 9
Grob, 2006, Cross- Population: volunteers, 107 X-ray: lateral Neck pain Unpaired There is no association Not provided 1
Switzerland33 sectional N45 y, visiting an projection of evaluated by t test between the sagittal
outpatient clinic at a Response cervical spine; questionnaire Pearson alignment of the cervical 1
hospital due to orthopedic rate: 73% measure-method: correlation spine and the presence
problems of posterior tangent coefficients of neck pain
the lower extremities; method; C2-C7 χ2 test 1
exclusion of persons with Fisher exact 0
previous neck trauma, test
3
ungoing treatment of
neck pain, or systemic
disease involving
the neck
Selection: NR
Kado, 2007, Cross- Population: 1883 Distance from Risk of injurious Logistic Simple clinical Summation 0
USA14 sectional participants from occiput to table in falls by regression observation of OR:
the Rancho supine position questionnaire analysis of being
Bernardo heart Response with head in χ2 test unable to lie Range of OR, 1
and chronic rate: 80% neutral position T test flat with the 1.59-1.93
disease study head positioned
established in within a
1972, which were margin of
alive in 1988- 3.4 cm from
1992 the examination
Selection: table could Range of 0
NR potentially aid in CI = 0.96-3.10
identifying older ORs for 1
men at high risk association
2
for future between
injurious falls; injurious
such correlations falls and
were not found kyphosis
for women
Lopes, 2007, Case- Population: 60 boys (20+20)/20 Occiput-to-wall Asthma diagnosis 1-way Unclear Not provided 1
Brasil34 control age 7-12 distance; standing and severity of ANOVA.
Cases: 20 cases with Response position asthma according 1
mild persistent asthma rate: NR to Global 1
and 20 cases with Initiative for
severe persistent asthma Asthma guidelines
Controls: sex and 0
age-matched; no atopic
3
disorder or respiratory
disease
LBP indicates low back pain; NR, not reported; ANOVA, analysis of variance; BMI, body mass index; MRI, magnetic resonance imaging.

707
Table 2. Summary of measurement methods used to assess sagittal spinal curves

Instrument /technology used Method No. of studies and references
By eye By eye 522,30,31,52,60
Distance From occiput to table 414,23,62,67
From occiput to wall 134
From occiput to wall and difference between SS and KHS 120
From wall to apex C7 130
Diverse measure instruments Inclinometer 328,50,56

Debrunner's kyphometer 166
Flexicurve, method of Conner and Breen 142
Flexible ruler, method of Youdas 226,27
Flexible rod (Milne and Lauder) 124
Spinal pantograph 216,64
Surveyor's flexicurve, (I/K) 219,63
3Space Isotrak device 158
MRI Supine position with cushings under knees 148

Photographs Cervicothoracic angle 154
Lateral posture 160
Method of Flint 136
X-ray Cervical spine 138

Cobb or Cobb-like method 1218,21,32,41,43-46,51,53,55,57
Farfans method 129
Instantaneous shape of lordotic curvature 137
Ishihara's method 149
Kyphotic angle 159
Lateral projection 217,35
Measured with a goniometer, Cobb-method 139
Posterior tangent method 325,33,41
Thorax, Ferguson's method 147
Thorax, index of wedging 163
Trigonometric method by Youdas 115
2-line Cobb method 140
No report of measure method 261,65

MRI indicates magnetic resonance imaging; SS, standing structure; KHS, knee height structure.
distance measurement (apex of curve to surface),14,20,23,30, Quality

34,62,67
and other specific-measure instruments15,16,19,24,26- Most studies (n = 30) obtained 2 points of the 3 or 4
28,42,50,56,58,63,64,66
(Table 2). In 6 studies, the sagittal possible. Details for each included article are provided in
curve of the cervical spine was evaluated25,33,38,41,49,54; in Table 1. Only 3 studies20,25,47 obtained maximum points and
34 studies, the thoracic sagittal curvature14,16,17,19,20,22- are considered to be of higher quality, Figure 1.
24,28,30,31,34-36,42-45,47,52-57,59-67
; and in 38 studies, the
lumbar sagittal curve.15-19,21,24,26-29,31,32,35-37,39,40,42-46,48,50-
53,55-58,60,63-66
. Level of Evidence
Altogether, 14 health outcomes were investigated in Results of the level of evidence analysis are also
relation to the spinal curves (Table 3). Of these, low back displayed in Table 3. According to our criteria, we did
pain was most commonly the focus (29 studies). Authors' not find strong evidence for association between the
conclusions regarding relationship between sagittal sagittal spinal curves and any of the health outcomes
curves and health outcomes are provided in Table 1. studied. We found moderate evidence for an association
In 3014,18,20,23-25,28,30-32,35,36,38,40-45,47,49,51-53,57-59,62,63,67 between the sagittal spinal curves and 4 health outcomes;
of the 54 studies, the authors concluded that there was a these were temporomandibular disorder,38,52 pelvic organ
positive association between the sagittal spinal curves and prolapse,24,47,51 daily function,23,28,30,57 and death.59,62,63
the health symptoms in question. The ratio of studies We found insufficient evidence for an association between
finding a positive vs a negative association between the sagittal spinal curves and all other health related outcomes
sagittal spinal curves and each health symptom is provided (low back pain, symptomatic degenerative lumbar disk
in Table 3. disease, disk herniation, cervical pain, thoracic pain, tension-
Table 3. Summation of evidence for associations between various health conditions and sagittal spinal curves in epidemiological studies
Health condition Total no. of studies/studies Studies of higher Level of evidence Strength of Prospective
with statistically significant quality reporting association* study*
positive associations positive associations (OR or RR)
Low back pain 29/1015,18,31,32,35,36,40,42,43,45,58 0 Insufficient evidence
Degenerative 1/144 0 Insufficient evidence
lumbar disc disease
Disk herniation 1/153 0 Insufficient evidence
Cervical pain 8/325,31,41 1 Insufficient evidence
OR, 18 (CI not reported)25
OR, 2 (CI not reported)25
Thoracic pain 2/0 0 Insufficient evidence
Tension type 2/149,61 0 Insufficient evidence
headache
Temporomandibular 2/238,52 0 Moderate evidence Not provided No
disorder
Pelvic organ prolapse 3/324,47,51 1 Moderate evidence OR, 3.18 (1.46-6.93)24 No
OR, 1.35 (1.11-1.65)47 OR, 1.35 (1.11-1.65)47
Death 3/359,62,63 0 Moderate evidence RH = 1.40 (1.08-1.81)62 Yes
Daily function 5/423,28,30,57 0 Moderate evidence OR, 1.64 (1.03-2.61)30 No
OR, 1.38 (1.12-1.68)30
Range of OR and CI,
1.8-3.7 (1.1-6.3)23
Bone mass loss 1/0 0 Insufficient evidence
Ventilatory 2/120,34 1 Insufficient evidence
dysfunction
OR, 2.5 (1.1-5.8) dyspnea20
OR, 3.3 (1.7-6.5)
obstructive ventilatory pattern20
OR, 2.3 (1.1-4.8)
restrictive ventilatory pattern20
Fracture 1/167 0 Insufficient evidence
Risk of injurious falls 1/114 0 Insufficient evidence
Displayed only for health outcomes with moderate or strong level of evidence for an association with spinal sagittal curves.
type headache, bone mass loss, ventilatory disorders, sagittal spinal curves and health. 58,59,62,63,67 Only a
fracture, risk of injurious falls). moderate or strong association was found between
thoracic hyperkyphosis and death, 62,63 but the increased
mortality in the hyperkyphotic groups could be explained
Strength of Associations by cardiovascular disease in both articles. For the other
Table 4 provides the strength of associations found outcomes, the positive associations did not reach statis-
between the spinal curves and the 18 health outcomes tical significance.
in the 13 studies where an OR or RR was repor-
ted.14,20,23-25,30,47,58,61,62,64,65,67 We considered only health
outcomes where a strong or moderate level of evidence for DISCUSSION
an association was found and these were as follows: pelvic
After reviewing and critically assessing 54 epidemiolo-
organ prolapse (moderate evidence, strong association [OR,
gical studies published before January 1, 2008, we
3.18; 95% CI, 1.46-96.9324] or moderate association [OR,
1.35; 95% CI, 1.11-1.6547]); daily function (moderate conclude that there is no evidence for a causal relationship
between sagittal spinal curvature and health including
evidence, strong association [OR-range, 1.8-3.7; 95% CI-
spinal pain. However, most studies did either not fulfill 4
range, 1.1-6.323] or moderate association [OR, 1.64; 95% CI,
1.03-2.6130; OR, 1.38; 95% CI, 1.12-1.6830]); and death very basic requirements for epidemiological studies or were
(moderate evidence, moderate association [OR, 1.40; 95% not prospective in design and thus not suitable for
CI, 1.08-1.8162]). addressing the fundamental issue of a possible causal
connection. Furthermore, in the 2 cases where a causal
connection may have existed, the finding was likely
Assessment of Causality and Confounding explained by confounding. We can however not entirely
Table 5 summarizes the prospective cohort studies rule out the possibility that there is a relationship between
where authors reported positive associations between the sagittal spinal curves and health, but on the basis of the
overall very low. Therefore, further detailing and tightening

of our quality criteria would likely have been futile and
resulted in even fewer studies being rated as higher quality
studies making overall assessment of the results even more
difficult. Nevertheless, 80% of the included studies fulfilled
the criterion for acceptable definition and validation of
method used to measure spinal sagittal curves (Table 1).
However, the operationalization of this item lead us to accept
any attempt at validation pertaining to either reproducibility
and/or validity and probably closer scrutiny of studies
specifically addressing this item would have revealed
deficient reproducibility and validity in many of the methods
used. Such in-depth scrutiny of individual methods was
outside the scope of this review.
Assessing LOE and strength of associations have been
used extensively in the literature.10-12 However, the
application of different LOE criteria has been shown to
influence conclusions of systematic reviews.9 We therefore
do not recommend a rigorous interpretation of the outcome
of such an analysis; however, we do consider an LOE
analysis to be one feasible way of summarizing information
Fig 1. Distribution of numbers of studies in relation to the especially if it is combined with other methods such as
quality score. assessment of strength of associations. This is because
statistical significance alone is also not very helpful in
results of this review, we question whether variations in understanding relationships between an exposure and an
sagittal spinal curves can be used to explain symptoms in outcome as this requires calculation and reporting of an OR
patients and/or to justify treatment. or an RR. Only in 13 of 54 included articles were such
Assessing causality in epidemiological studies requires measures of association reported. Thus we conclude that in
evaluation of several factors as outlined by Hill.13 Most the case of a rather large body of literature of low
importantly, the exposure needs to precede the effect. We methodological quality such as here, it is unlikely that the
did not find evidence for an association between spinal application of other summarizing tools would result in
curves and health in any of the prospective studies included radically different conclusions.
in this review that could not be explained by confounding According to the chosen criteria, we did not find any
or bias (see below). However, other factors such as relationships between spinal sagittal curves and health
consistency (repeated observation of an association in exhibiting a strong LOE for an association, that is, evidence
different populations under different circumstances), from multiple higher quality studies. In only 3 of 4 health-
strength of the associations (strong associations are more related outcomes (pelvic organ prolapse, 24,47 daily
likely to be causal than weak ones), biologic gradient (the function,23,30 and death62) exhibiting a moderate LOE for
presence of a monotonic dose-response curve), and an association were an OR or an RR reported. Of these, only
plausibility of a causal connection are also important and the outcome death was assessed in a prospective cohort study.
should be considered.13 Although it is biologically plausible However, in spite of a statistically significant association, the
that a causal connection exists between spinal curves and mortality in the study could be explained by death due to
for instance back or neck pain,1 we found inconsistent cardiovascular disease and not by abnormal spinal curves.62
conclusions between the articles, weak associations, and no The results of this review must be interpreted in the light
evidence for a dose-response relationship, that is, greater of several weaknesses. First and foremost, locating all
curves did not equal more health problems. relevant studies within the area of interest is probably
Assessment of possible biases is essential in systematic impossible. Producing very specific and narrow searches in
reviews, and such biases may be addressed by systematic the electronic databases were difficult because many
attention to factors pertaining to participants (composition of different keywords relate to the topic of this review. Thus,
the study sample), loss to follow-up, definition and our searches produced a total of 11 949 potentially relevant
measurement of the factor of interest, the outcome of citations that were subsequently screened for possible
interest, confounding and statistical analysis in the individual inclusion by the first author. Studies that had a main focus
studies.68 We operationalized these issues using our quality on the spinal curvature in relation to health were easy to
score that addressed all of these at a very basic level and identify, and we believe all such studies were identified and
found that the methodological quality of the literature was included. However, in many articles, substudies relating to
Table 4. Strength of associations between sagittal spinal curvatures and various health conditions in studies where OR or RR is reported
Sagittal spinal curvature and health condition (reference) OR/RR
20
Strong association Kyphosis and obstructive ventilatory pattern OR, 3.3 (1.7-6.5)
Abnormal spinal curvature and pelvic organ prolapse24 OR, 3.18 (1.46-6.93)
Kyphosis and dyspnea20 OR, 2.5 (1.1-5.8)
Kyphosis and restrictive ventilatory pattern20 OR, 2.3 (1.1-4.8)
Kyphosis and poor physical function23 Summation of OR:
Range of OR = 1.8-3.7
Range of CI = 1.1-6.3
Moderate association Reduced lordosis and LBP58 OR, 1.8 (1.0-3.2)

Hyperkyphosis and future fracture67 Summation of OR:
Range of OR, 1.73-1.77
Range of CI, 1.02-3.05
Qualitative measurement of kyphosis and heavy household30 OR, 1.64 (1.03-2.61)
Kyphosis and risk of injurious falls14 Summation of OR:
Range of OR = 1.59-1.93
Range of CI = 0.96-3.10
Relative hazard ratio for hyper kyphotic posture in relation to death62 Relative hazard = 1.40 (1.08-1.81)
Quantitave measurements of kyphosis and heavy household30 OR, 1.38 (1.12-1.68)
Each additional degree of kyphosis and uterine prolapse47 OR, 1.35 (1.11-1.65)
No association Kyphosis or lordosis and LBP65 Summation of ORs:

Range of OR = 0.87-1.22
Range of CI = 0.63-1.69
Kyphosis or lordosis and LBP64 Summation of ORs:
Range of OR = 0.84-1.20
Range of CI = 0.61-1.50
Reduced cervical lordosis and cervical pain25 OR, 18 (CI = not reported)
Reduced cervical lordosis and cervical pain25 OR, 2 (CI = not reported)
RR for cervical pain in relation to posture deviation61 RR = 1.5 (CI = not reported)
RR for lumbar pain in relation to posture61 1.0 (nonsignificant)
LBP indicates low back pain.
the issue at hand were nested in larger clinical studies, and Table 5. Longitudinal studies with positive association between
such studies proved more difficult to locate. We (both sagittal spinal curves and health
authors) meticulously read abstracts and obtained full articles Health condition (reference) Sagittal curvature Quality score and OR
whenever we suspected that this was the case, but there may Low back pain58 Reduced lumbar 3/4
still be several smaller studies dealing with these issues out lordosis OR, 1.8 (1.0-3.2)
there. In spite of this shortcoming, we consider this the most Death62 Thoracic 2/4
comprehensive and systematic review of spinal sagittal hyperkyphosis RH, 1.40 (1.08-1.81)
Death63 Thoracic 2/4
curves in relation to health published so far. hyperkyphosis OR not reported
We based our quality scoring system on 4 basic domains Fracture of Thoracic 2/4
related to our research question called participants, exposure, thoracolumbar spine67 hyperkyphosis OR, 1.73-1.77 (1.02-3.05)
outcome, and analysis. Furthermore, the items were Survival time59 Thoracic 1/4
operationalized based on recommendations for this type of hyperkyphosis OR not reported
review.68 It is important for us to emphasize that the results
of the evaluation of individual studies should not be
interpreted in a rigorous fashion. Rather, it is the consistent Previously, authors of nonsystematic reviews dealing
message from studies in a variety of settings applying with this subject have arrived at conflicting conclusions.1-5,7
different definitions of exposures in relation to different After reviewing the rather extensive epidemiological
outcomes that can only be interpreted as a lack of literature dealing with the relationship between sagittal
documented exposure-effect relationship between spinal spinal curves and health, we find no evidence for an
sagittal curves and health in general. The operational criteria association. However, we cannot rule out that such a
for obtaining points were set quite low, and we do not believe relationship may exist because the overall methodological
that a more detailed scoring system would have changed the quality of the published studies was quite low. New studies
overall conclusions although it may have revealed interesting using improved methodology are therefore likely to affect
details about individual studies. these conclusions.
We recommend that clinicians accept these conclusions 4. Gay RE. The curve of the cervical spine: variations and
and continue to be aware of new and hopefully better studies significance. J Manipulative Physiol Ther 1993;16:591-4.
5. Ames RA. Posture in the assessment, diagnosis and treatment
investigating these issues. It is important to keep in mind that of chronic low back pain. J Aust Chiropr Assoc 1985;15:
spinal curves may still play a role in pain syndromes of 21-31.
individual patients, and it would probably be unwise to 6. Paulk GP, Harrison DE. Management of a chronic lumbar disk
dismiss a connection between spinal curves and health in all herniation with chiropractic biophysics methods after failed
patients. Nevertheless, basing treatment regimens on the chiropractic manipulative intervention. J Manipulative Physiol
Ther 2004;27:579.
sagittal configuration of the spine appears to be counter- 7. Haas M, Taylor JA, Gillette RG. The routine use of
intuitive at this point. For researchers, the challenge is radiographic spinal displacement analysis: a dissent.
determine exactly what constitutes normal spinal sagittal J Manipulative Physiol Ther 1999;22:254-9.
curves and to put forward studies with valid measurements of 8. von Elm E, Altman DG, Egger M, Pocock SJ, Goetzshe PC,
spinal curves and health and apply these in prospective Vandenbroucke JP. The STrengthening of the Reporting of
Observational studies in Epidemiology (STROBE) statement:
cohorts to study who develops pain and ill health and whether guidelines for reporting observational studies. J Clin Epidemiol
this is dependent on the sagittal configuration of their spine. 2008;61:344-9.
9. Ferreira PH, Ferreira ML, Maher CG, Refshauge K, Herbert
RD, Latimer J. Effect of applying different “levels of evidence”
CONCLUSION criteria on conclusions of Cochrane reviews of interventions for
low back pain. J Clin Epidemiol 2002;55:1126-9.
After systematically reviewing the epidemiological litera- 10. Hartvigsen J, Lings S, Leboeuf-Yde C, Bakketeig L.
ture associated with sagittal spinal curves in relation to health, Psychosocial factors at work in relation to low back pain and
consequences of low back pain; a systematic, critical review of
we found no evidence that a relationship exists. However, prospective cohort studies. Occup Environ Med 2004;61:e2.
most studies either did not fulfill basic requirements for 11. van Tulder M, Furlan A, Bombardier C, Bouter L. Updated
epidemiological studies or were not prospective in design. We method guidelines for systematic reviews in the Cochrane
therefore encourage researchers to produce new and better collaboration back review group. Spine 2003;28:1290-9.
studies to determine the role sagittal spinal curves play in 12. Hemingway H, Marmot M. Evidence based cardiology:
psychosocial factors in the aetiology and prognosis of coronary
health and disease. At present, it is important for clinicians to heart disease. Systematic review of prospective cohort studies.
know when communicating with patients that sagittal spinal BMJ 1999;318:1460-7.
curves do not appear to be associated with health or pain. 13. Hill AB. The environment and disease: association or
causation. Proc R Soc Med 1965;58:295-300.
14. Kado DM, Huang MH, Nguyen CB, Barrett-Connor E,
Greendale GA. Hyperkyphotic posture and risk of injurious
falls in older persons: the Rancho Bernardo Study. J Gerontol A
Practical Applications
Biol Sci Med Sci 2007;62:652-7.
15. Norton BJ, Sahrmann SA, Van Dillen FL. Differences in
• Results from epidemiological studies published measurements of lumbar curvature related to gender and low
between 1942 and 2008 does not support that an back pain. J Orthop Sports Phys Ther 2004;34:524-34.
association between sagittal spinal curves and 16. Bergenudd H, Nilsson B, Uden A, Willner S. Bone mineral
content, gender, body posture, and build in relation to back pain
health, including spinal pain. in middle age. Spine 1989;14:577-9.
• More studies of better quality are needed before 17. Biering-Sorensen F, Hansen FR, Schroll M, Runeborg O.
conclusions about association can be drawn. The relation of spinal x-ray to low-back pain and physical
• Clinicians should downplay the importance of activity among 60-year-old men and women. Spine 1985;10:
sagittal spinal curves in relation to health until 445-51.
18. Bryner P. Lumbar spine lordosis in low-back pain: an analysis
more evidence for the importance is available. of radiographs. Chiropr J Aust 1992;22:42-6.
19. Chow RK, Harrison JE. Relationship of kyphosis to physical
fitness and bone mass on post-menopausal women. Am J Phys
REFERENCES Med 1987;66:219-27.
20. Di Bari M, Chiarlone M, Matteuzzi D, et al. Thoracic kyphosis
1. Troyanovich SJ, Harrison DE, Harrison DD. Structural and ventilatory dysfunction in unselected older persons: an
rehabilitation of the spine and posture: rationale for treatment epidemiological study in Dicomano, Italy. J Am Geriatr Soc
beyond the resolution of symptoms. J Manipulative Physiol 2004;52:909-15.
Ther 1998;21:37-50. 21. Frymoyer JW, Newberg A, Pope MH, Wilder DG, Clements J,
2. Harrison DE, Harrison DD, Troyanovich SJ. Reliability of MacPherson B. Spine radiographs in patients with low-back
spinal displacement analysis of plain x-rays: a review of pain. An epidemiological study in men. J Bone Joint Surg Am
commonly accepted facts and fallacies with implications for 1984;66:1048-55.
chiropractic education and technique. J Manipulative Physiol 22. Griegel-Morris P, Larson K, Mueller-Klaus K, Oatis CA.
Ther 1998;21:252-66. Incidence of common postural abnormalities in the cervical,
3. Balague F, Troussier B, Salminen JJ. Non-specific low back shoulder, and thoracic regions and their association with pain
pain in children and adolescents: risk factors. Eur Spine J 1999; in two age groups of healthy subjects. Phys Ther 1992;72:
8:429-38. 425-31.
23. Kado DM, Huang MH, Barrett-Connor E, Greendale GA. with low back pain matched for age, sex, and size. A
Hyperkyphotic posture and poor physical functional ability in prospective controlled clinical study. Spine 1994;19:1611-8.
older community-dwelling men and women: the Rancho 44. Jackson RP, Peterson MD, McManus AC, Hales C. Compen-
Bernardo study. J Gerontol A Biol Sci Med Sci 2005;60: satory spinopelvic balance over the hip axis and better
633-7. reliability in measuring lordosis to the pelvic radius on standing
24. Mattox TF, Lucente V, McIntyre P, Miklos JR, Tomezsko J. lateral radiographs of adult volunteers and patients. Spine 1998;
Abnormal spinal curvature and its relationship to pelvic organ 23:1750-67.
prolapse. Am J Obstet Gynecol 2000;183:1381-4. 45. Korovessis P, Stamatakis M, Baikousis A. Segmental roent-
25. McAviney J, Schulz D, Bock R, Harrison DE, Holland B. genographic analysis of vertebral inclination on sagittal plane
Determining the relationship between cervical lordosis and in asymptomatic versus chronic low back pain patients. J Spinal
neck complaints. J Manipulative Physiol Ther 2005;28: Disord 1999;12:131-7.
187-93. 46. Korovessis P, Dimas A, Iliopoulos P, Lambiris E. Correlative
26. Nourbakhsh MR, Moussavi SJ, Salavati M. Effects of lifestyle analysis of lateral vertebral radiographic variables and medical
and work-related physical activity on the degree of lumbar outcomes study short-form health survey: a comparative study
lordosis and chronic low back pain in a Middle East population. in asymptomatic volunteers versus patients with low back pain.
J Spinal Disord 2001;14:283-92. J Spinal Disord Tech 2002;15:384-90.
27. Nourbakhsh MR, Arab AM. Relationship between mechanical 47. Lind LR, Lucente V, Kohn N. Thoracic kyphosis and the
factors and incidence of low back pain. J Orthop Sports Phys prevalence of advanced uterine prolapse. Obstet Gynecol 1996;
Ther 2002;32:447-60. 87:605-9.
28. O'Brien K, Culham E, Pickles B. Balance and skeletal 48. Murrie VL, Dixon AK, Hollingworth W, Wilson H, Doyle TA.
alignment in a group of elderly female fallers and nonfallers. Lumbar lordosis: study of patients with and without low back
J Gerontol A Biol Sci Med Sci 1997;52:B221-6. pain. Clin Anat 2003;16:144-7.
29. Pope MH, Bevins T, Wilder DG, Frymoyer JW. The relation- 49. Nagasawa A, Sakakibara T, Takahashi A. Roentgenographic
ship between anthropometric, postural, muscular, and mobility findings of the cervical spine in tension-type headache.
characteristics of males ages 18-55. Spine 1985;10:644-8. Headache 1993;33:90-5.
30. Ryan SD, Fried LP. The impact of kyphosis on daily 50. Ng JK, Richardson CA, Kippers V, Parnianpour M. Comparison
functioning. J Am Geriatr Soc 1997;45:1479-86. of lumbar range of movement and lumbar lordosis in back pain
31. Salminen JJ. The adolescent back. A field survey of 370 finish patients and matched controls. J Rehabil Med 2002;34:109-13.
school children. Acta Peadiatrica Scandinavia 1984;315:1-122. 51. Nguyen JK, Lind LR, Choe JY, McKindsey F, Sinow R, Bhatia
32. Tsuji T, Matsuyama Y, Sato K, Hasegawa Y, Yimin Y, Iwata H. NN. Lumbosacral spine and pelvic inlet changes associated
Epidemiology of low back pain in the elderly: correlation with with pelvic organ prolapse. Obstet Gynecol 2000;95:332-6.
lumbar lordosis. J Orthop Sci 2001;6:307-11. 52. Nicolakis P, Nicolakis M, Piehslinger E, et al. Relationship
33. Grob D, Frauenfelder H, Mannion AF. The association between between craniomandibular disorders and poor posture. Cranio
cervical spine curvature and neck pain. Eur Spine J 2006;16: 2000;18:106-12.
669-78. 53. Rajnics P, Templier A, Skalli W, Lavaste F, Illes T. The
34. Lopes EA, Fanelli-Galvani A, Prisco CC, et al. Assessment of importance of spinopelvic parameters in patients with lumbar
muscle shortening and static posture in children with persistent disc lesions. Int Orthop 2002;26:104-8.
asthma. Eur J Pediatr 2007;166:715-21. 54. Refshauge K. The relationship between cervicothoracic posture
35. Brav EA. A roentgenologic study of low back and sciatic pain. and the presence of pain. J Manipulative Physiol Ther 1995;3:
AJR Am J Roentgenol 1942;48:39-46. 21-4.
36. Christie HJ, Kumar S, Warren SA. Postural aberrations in low 55. Tuzun C, Yorulmaz I, Cindas A, Vatan S. Low back pain and
back pain. Arch Phys Med Rehabil 1995;76:218-24. posture. Clin Rheumatol 1999;18:308-12.
37. During J, Goudfrooij H, Keessen W, Beeker TW, Crowe A. 56. Waddell G, Somerville D, Henderson I, Newton M. Objective
Toward standards for posture. Postural characteristics of the clinical evaluation of physical impairment in chronic low back
lower back system in normal and pathologic conditions. Spine pain. Spine 1992;17:617-28.
1985;10:83-7. 57. Sinaki M, Brey RH, Hughes CA, Larson DR, Kaufman KR.
38. Evcik D. Relationship between head posture and temporoman- Balance disorder and increased risk of falls in osteoporosis and
dibular dysfunction syndrome. J Musculoskelet Pain 2004;12: kyphosis: significance of kyphotic posture and muscle strength.
19-24. Osteoporos Int 2005;16:1004-10.
39. Hansson T, Bigos S, Beecher P, Wortley M. The lumbar lordosis 58. Adams MA, Mannion AF, Dolan P. Personal risk factors for
in acute and chronic low-back pain. Spine 1985;10:154-5. first-time low back pain. Spine 1999;24:2497-505.
40. Harrison DD, Cailliet R, Janik TJ, Troyanovich SJ, Harrison 59. Anderson F, Cowan NR. Survival of healthy older people. Br J
DE, Holland B. Elliptical modeling of the sagittal lumbar Prev Soc Med 1976;30:231-2.
lordosis and segmental rotation angles as a method to 60. Dieck GS, Kelsey JL, Goel VK, Panjabi MM, Walter SD,
discriminate between normal and low back pain subjects. J Laprade MH. An epidemiologic study of the relationship
Spinal Disord 1998;11:430-9. between postural asymmetry in the teen years and subsequent
41. Harrison DD, Harrison DE, Janik TJ, et al. Modeling of the back and neck pain. Spine 1985;10:872-7.
sagittal cervical spine as a method to discriminate hypolordosis: 61. Hertzberg A. Prediction of cervical and low-back pain based on
results of elliptical and circular modeling in 72 asymptomatic routine school health examinations. A nine- to twelve-year
subjects, 52 acute neck pain subjects, and 70 chronic neck pain follow-up study. Scand J Prim Health Care 1985;3:247-53.
subjects. Spine 2004;29:2485-92. 62. Kado DM, Huang MH, Karlamangla AS, Barrett-Connor E,
42. Hein T. Sagittal spinal posture in subjects with and without Greendale GA. Hyperkyphotic posture predicts mortality in
back pain. Thesis Anglo-European college of Chiropractic, older community-dwelling men and women: a prospective
Bournemouth, England 1998. study. J Am Geriatr Soc 2004;52:1662-7.
43. Jackson RP, McManus AC. Radiographic analysis of sagittal 63. Milne JS, Williamson J. A longitudinal study of kyphosis in
plane alignment and balance in standing volunteers and patients older people. Age Ageing 1983;12:225-33.
64. Nissinen M, Heliovaara M, Seitsamo J, Alaranta H, Poussa M. 66. Widhe T. Spine: posture, mobility and pain. A longitudinal
Anthropometric measurements and the incidence of low back study from childhood to adolescence. Eur Spine J 2001;10:
pain in a cohort of pubertal children. Spine 1994;19:1367-70. 118-23.
65. Poussa MS, Heliovaara MM, Seitsamo JT, Kononen MH, 67. Huang MH, Barrett-Connor E, Greendale GA, Kado DM.
Hurmerinta KA, Nissinen MJ. Anthropometric measurements Hyperkyphotic posture and risk of future osteoporotic fractures:
and growth as predictors of low-back pain: a cohort study of the Rancho Bernardo study. J Bone Miner Res 2006;21:419-23.
children followed up from the age of 11 to 22 years. Eur Spine J 68. Hayden JA. Evaluation of the quality of prognosis studies in
2005;14:595-8. systematic reviews. Ann Intern Med 2006;144:427-37.

SPINAL CURVES AND HEALTH A SYSTEMATIC CRITICAL

Uploaded by

Copyright:

Available Formats

SPINAL CURVES AND HEALTH A SYSTEMATIC CRITICAL

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

SPINAL CURVES AND HEALTH A SYSTEMATIC CRITICAL

Uploaded by

Copyright:

Available Formats

LITERATURE REVIEW

SPINAL CURVES AND HEALTH: A SYSTEMATIC CRITICAL

ome health care providers believe that sagittal spinal

Volume 31, Number 9

Christensen and Hartvigsen

Spinal Curves and Health

Journal of Manipulative and Physiological Therapeutics

Christensen and Hartvigsen

Spinal Curves and Health

Spinal Curves and Health

Journal of Manipulative and Physiological Therapeutics

Christensen and Hartvigsen

Spinal Curves and Health

Spinal Curves and Health

Journal of Manipulative and Physiological Therapeutics

Christensen and Hartvigsen

Spinal Curves and Health

Journal of Manipulative and Physiological Therapeutics

Christensen and Hartvigsen

Spinal Curves and Health

Spinal Curves and Health

Journal of Manipulative and Physiological Therapeutics

Christensen and Hartvigsen

Spinal Curves and Health

Spinal Curves and Health

Journal of Manipulative and Physiological Therapeutics

Christensen and Hartvigsen

Spinal Curves and Health

Journal of Manipulative and Physiological Therapeutics

Christensen and Hartvigsen

Table 2. Summary of measurement methods used to assess sagittal spinal curves

Diverse measure instruments Inclinometer 328,50,56

MRI Supine position with cushings under knees 148

X-ray Cervical spine 138

No report of measure method 261,65

distance measurement (apex of curve to surface),14,20,23,30, Quality

overall very low. Therefore, further detailing and tightening

Moderate association Reduced lordosis and LBP58 OR, 1.8 (1.0-3.2)

No association Kyphosis or lordosis and LBP65 Summation of ORs:

You might also like