International Journal of COPD Dovepress

open access to scientific and medical research

Open Access Full Text Article Original Research

Low FEV1, smoking history, and obesity are factors

associated with oxygen saturation decrease in an
adult population cohort
This article was published in the following Dove Press journal:
International Journal of COPD
21 October 2014
Number of times this article has been viewed

Monica Linea Vold 1,3 Background: Worsening of pulmonary diseases is associated with a decrease in oxygen
Ulf Aasebø 1,2 ­saturation (SpO2). Such a decrease in SpO2 and associated factors has not been previously
Hasse Melbye³ evaluated in a general adult population.
Aim: We sought to describe SpO2 in a sample of adults, at baseline and after 6.3 years, to
1
Department of Respiratory Medicine,
University Hospital of North Norway, determine whether factors predicting low SpO2 in a cross-sectional study were also associated
2
Department of Clinical Medicine, with a decrease in SpO2 in this cohort.
3
Department of Community Medicine,
University of Tromsø, Tromsø, Norway
Methods: As part of the Tromsø Study, 2,822 participants were examined with pulse oximetry in
Tromsø 5 (2001/2002) and Tromsø 6 (2007/2008). Low SpO2 by pulse oximetry was defined as an
SpO2 #95%, and SpO2 decrease was defined as a $2% decrease from baseline to below 96%.
Results: A total of 139 (4.9%) subjects had a decrease in SpO 2. Forced expiratory vol-
ume in 1 second (FEV1) ,50% of the predicted value and current smoking with a his-
tory of $10 pack-years were the baseline characteristics most strongly associated with
an SpO2 decrease in multivariable logistic regression (odds ratio 3.55 [95% confidence
interval (CI) 1.60–7.89] and 2.48 [95% CI 1.48–4.15], respectively). Male sex, age,
former smoking with a history of $10 pack-years, body mass index $30 kg/m2, and
C-reactive protein $5 mg/L were also significantly associated with an SpO2 decrease.
A significant decrease in FEV1 and a new diagnosis of asthma or chronic obstructive pulmonary
disease during the observation period most strongly predicted a fall in SpO2. A lower SpO2
decrease was observed in those who quit smoking and those who lost weight, but these tenden-
cies were not statistically significant.
Conclusion: A decrease in SpO2 was most strongly associated with severe airflow limitation and
a history of smoking. Smoking cessation and reducing obesity seem to be important measures
to target for avoiding SpO2 decreases in the general population.
Keywords: pulse oximetry, lung function, cohort study, general population

Introduction
Pulse oximetry is an inexpensive, noninvasive method for measuring oxygen saturation
(SpO2). Pulse oximetry has a wide range of use both in primary pulmonary care and
critical care medicine. Low SpO2/hypoxemia have been associated with conditions
or diseases causing ventilation–perfusion mismatch in the lungs, hypoventilation,
Correspondence: Monica Linea Vold right-to-left shunts, reduced diffusion capacity, and reduced oxygen partial pressure
Department of Respiratory Medicine,
University Hospital of North Norway, in inspired air. Decrease in SpO2/desaturation has been associated with the worsening
9038 Tromsø, Norway of preexisting pulmonary diseases.1–3
Tel +47 776 26828
Fax +47 776 28261
There is no clear cutoff point for abnormal SpO2, but SpO2 #95% is used in most
Email [email protected] adult studies. In their blood gas reference values for sea level, Crapo et al found mean

submit your manuscript | www.dovepress.com International Journal of COPD 2014:9 1225–1233 1225
Dovepress © 2014 Vold et al. This work is published by Dove Medical Press Limited, and licensed under Creative Commons Attribution – Non Commercial (unported, v3.0)
http://dx.doi.org/10.2147/COPD.S69438
License. The full terms of the License are available at http://creativecommons.org/licenses/by-nc/3.0/. Non-commercial uses of the work are permitted without any further
permission from Dove Medical Press Limited, provided the work is properly attributed. Permissions beyond the scope of the License are administered by Dove Medical Press Limited. Information on
how to request permission may be found at: http://www.dovepress.com/permissions.php
Vold et al Dovepress

arterial oxygen saturation (SaO2) to be 95.5%–96.9% (stan- an extra, more extended medical examination; a total of 7,916
dard deviation [SD] 0.4%–1.4%), depending on age.4 Resting (77%) participated. All participants who had this second visit
SpO2 #95% has been found to predict oxygen desaturation in Tromsø 4 were invited to the Tromsø 5 Study21 and were
during sleep, exercise, and flights, in chronic obstructive again eligible for a second, extended, medical examination.
pulmonary disease (COPD) patients.5–7 SpO2 #95% has also As part of the fifth Tromsø Study (2001/2002), 5,152 subjects
been identified as a risk factor for postoperative pulmonary were examined with pulse oximetry. Of these, 3,453 (67.0%)
complications.8 The limit of 96% therefore seems a reason- participants also took part in Tromsø 6 (2007/2008),21 and
able cutoff value. A cutoff value of #92% has been used 3,127 (60.7%) attended the extended examination. Figure 2
when screening for respiratory failure in COPD.9 shows the flow chart of participants from Tromsø 5 to 6.
In a previous cross-sectional study, we have shown that A total of 9.8% of the participants were not examined
body mass index (BMI) and the forced expiratory volume with pulse oximetry and spirometry due to absence of staff or
in 1 second (FEV1) as a percentage of the predicted value drop out related to wait time for lung function testing. SpO2
(FEV1 % predicted) are the most important predictors of low values of 2,822 participants were measured in both Tromsø
SpO2 in the general adult population.10 Other predictors for 5 and Tromsø 6. The mean time between measurements was
low SpO2 are former and current smoking, C-reactive protein 6.3 years (SD 0.4 years).
(CRP) $5 mg/L, age, male sex, elevated hemoglobin, and
respiratory symptoms. Examinations
The role of inflammation in lung function decline is not In both Tromsø 5 and Tromsø 6, a questionnaire including
clearly understood. CRP and other biomarkers have, in COPD, medical history and smoking habits was enclosed in the invi-
been associated with progression of the disease and decline tation to participate. Participants who reported suffering from
in lung function.11–13 Systemic inflammation in COPD might angina pectoris, myocardial infarction, or cerebral stroke were
play a role in the development of extrapulmonary comorbid classified as “self-reported CVD”. “Pack-years” of cigarette
conditions.14,15 Elevated CRP levels have previously been use was calculated by multiplying the average number of
found to be associated with cardiovascular disease (CVD), cigarettes smoked daily by the number of years smoked and
metabolic syndrome, and obesity. In sleep apnea, elevated dividing the product by 20. Subjects who attended the assess-
CRP has been associated with hypoxemia.16,17 Sleep apnea ment received an additional questionnaire about dyspnea,
is associated with obesity and metabolic ­syndrome, both cough, and sputum. Examinations at the first visit included
characterized by systemic inflammation and ­comorbidities.18 height and weight, and BMI (kg/m2) was calculated.
In some studies, elevated CRP has been associated with Pulse oximetry and spirometry were included at the
hypoxemia in COPD patients.19,20 second visit for both Tromsø 5 and Tromsø 6. SpO2 values
Lung function decline in adult population cohorts has were measured with a digital handheld pulse oximeter (Onyx
been evaluated by spirometry, but decrease in SpO2 has not II® 9550; Nonin Medical, Inc., Plymouth, MN, USA). Par-
been studied. We wanted to investigate changes in SpO2 in ticipants rested at least 15 minutes before examination. The
an adult population cohort to determine whether parameters best of three measurements was recorded. The manufac-
predicting low SpO2 in a cross-sectional study were also turer’s testing has shown that only values between 70% and
associated with a decrease in SpO2 in a cohort study. 100% are accurate to within ±2%, and values below 70%
are regarded as invalid. None of the participants received
Material and methods supplemental oxygen.
Subjects Spirometry was carried out using a SensorMedics Vmax™
A cohort of the adult population in Tromsø, Norway has Legacy 20® (VIASYS Healthcare Respiratory ­Technologies,
been followed in the Tromsø Study since 1974. Tromsø is Yorba Linda, CA, USA) in Tromsø 5, and the Vmax Encore
a university city in northern Norway, with approximately 20® (VIASYS Healthcare Respiratory Technologies) in
70,000 inhabitants. To date, the Tromsø Study has consisted Tromsø 6. American Thoracic Society (ATS)/European
of six cross-sectional studies. Participant selection in Tromsø Respiratory Society (ERS) criteria for spirometry testing
4 (1994/1995) has influenced later studies as described in the were followed.22 ­Norwegian reference values for prebron-
cohort profile (Figure 1).21 In the fourth study, all inhabitants chodilator spirometry were used because reversibility testing
of Tromsø 55–74 years of age, and 5%–10% of the samples in was not performed.23 Three trained technicians conducted
the other cohorts aged 25–84 years were asked to take part in the spirometry.

1226 submit your manuscript | www.dovepress.com International Journal of COPD 2014:9

Dovepress
Dovepress Factors associated with oxygen saturation decrease

26,956 attended first visit T4

10,542 eligible for second visit T4

2,626 did not attend for second visit T4

7,916 attended second visit T4

533 died

361 moved/emigrated

7,022 eligible for follow-up

1,117 did not attend second visit T5

5,905 attended second visit T5

753 not examined with pulse oximetry

5,152 examined with pulse oximetry

Figure 1 Participation from Tromsø 4 (T4) to Tromsø 5 (T5).

Adjustment of spirometry results mean FEV1 value found with the Vmax Legacy was 2.5%
The mean annual FEV1 decline of 14 mL/year (standard (66 mL [SE 14 mL]) lower than that measured by the
error [SE] 0.8) was lower than expected. The lowest mean Vmax Encore. We therefore chose to correct the FEV1
decline recorded, in samples of women who never smoked, values in Tromsø 5 by adding 2.5%. ­Likewise, forced
was 17.6 mL/year, but higher values, depending on sex, vital capacity (FVC), was 5.2% (188 mL [SE 25 mL])
age, and history of smoking, are usually found.24,25 We lower when the Vmax Legacy was used compared with
therefore considered potential sources of bias. The use of the Vmax Encore.
two different spirometers in Tromsø 5 and 6 was a likely
source. The Norwegian supplier confirmed that the Vmax Laboratory samples
Legacy used in Tromsø 5 probably provided values that Blood was drawn for high-sensitivity CRP, fibrinogen, and
were too low and that this was not the case for Vmax uric acid analyses (also biomarkers of inflammation). For 3
Encore used in Tromsø 6, but no documentation could consecutive days, albumin and creatinine were measured in
be provided. Küenzli et al have demonstrated that using urine, and the albumin:creatinine ratio (ACR) was estimated
different spirometers in longitudinal studies is a source for each day. Mean values were used in the analysis, and an
of bias. 26 We therefore tested 48 subjects, 24 patients ACR between 3.0 and 30.0 mg/mmol was used as an indica-
and 24 voluntary employees using both spirometers. The tion of microalbuminuria.

International Journal of COPD 2014:9 submit your manuscript | www.dovepress.com

1227
Dovepress
Vold et al Dovepress

5,152 examined with pulse oximetry T5

485 died

153 moved/emigrated
4,514 eligible T6

1,061 did not attend T6

3,453 attended first visit T6

326 did not attend second visit T6

3,127 attended second visit T6

305 not examined with pulse oximetry T6

2,822 examined pulse oximetry T6

Figure 2 Participation from Tromsø 5 (T5) to Tromsø 6 (T6).

Statistical analysis Results

Low SpO2 was defined by SpO2 #95%, and SpO2 decrease Among the 2,822 participants who had SpO2 measured in
was defined by $2% decrease from baseline to ,96%. Diff­ both studies, 241 (8.5%) and 213 (7.5%) had an SpO2#95%
erences in continuous variables between subgroups at baseline in Tromsø 5 and Tromsø 6, respectively. Fifty-six (2.0%) had
(Tromsø 5) were explored using the Mann–Whitney U test, SpO2 #95% in both studies. Sixteen (0.6%) in Tromsø 5
and differences found between baseline and after 6.3 years and 25 (0.9%) participants in Tromsø 6 had SpO2 #92%.
were explored using the Wilcoxon signed-rank test. ­Frequency SpO2 decrease was seen in 139 (4.9%) participants. Mean
of SpO2 decrease was analyzed by sex, age, ­smoking habit, age at baseline of the 2,822 participants was 63.2 (SD 8.9)
spirometry, BMI, CRP, fibrinogen, uric acid, ACR, self-reported years, range 32–81 years. No significant difference was
health and diseases, and pulmonary symptoms. ­Continuous seen between mean SpO2 at baseline (97.4%) and that found
variables (age, BMI, FEV1, fibrinogen, uric acid, ACR, and after 6.3 years (97.3%). Self-reported diseases were more
CRP) were categorized, and the statistical significance of dif- frequently reported in Tromsø 6 than in Tromsø 5, but the
ferences was analyzed by chi-square test. ­Predictors of SpO2 frequency of current smoking dropped from 23.7 to 15.2
decrease with a statistical significance of ,5% were entered (Table 1). Valid spirometry was found in 2,728 participants,
into a multivariable binary logistic regression and excluded by for both baseline and after 6.3 years. Mean FEV1 % predicted
backward stepwise elimination. Only predictors with P,0.05 increased significantly, from 88.7% to 90.8% (P,0.001).
were kept in the final model. Changes from baseline variables There was no significant change in mean BMI, but in those
other than SpO2 were registered and continuous variables cat- with BMI $30 kg/m2 at baseline, the BMI decreased signifi-
egorized, based on one standard deviation. The associations cantly, from 32.8 to 32.4 (P,0.001).
with SpO2 decrease were analyzed by chi-square test. IBM Table 2 displays frequency of SpO2 decrease by baseline
SPSS Statistics for Windows, Version 21 (IBM Corp, Armonk, characteristics. Age, male sex, self-reported CVD, obesity
NY, USA) was used. (BMI $30 kg/m2), FEV1 % predicted, chronic cough with
The Regional Committee for Medical and Health sputum, and smoking were all significantly associated with a
Research Ethics in North Norway approved the Tromsø 5 and SpO2 decrease in univariable analysis, as were the biomarkers
6 surveys. All participants gave written, informed consent. CRP, fibrinogen, and uric acid.

1228 submit your manuscript | www.dovepress.com International Journal of COPD 2014:9

Dovepress
Dovepress Factors associated with oxygen saturation decrease

Table 1 Characteristics of 2,822 participants in Tromsø 5 Table 1 (Continued)

(baseline) and Tromsø 6 (6.3 years later) Tromsø 5 Tromsø 6
Tromsø 5 Tromsø 6 n (%) n (%)
n (%) n (%) SpO2
Sex $96 2,581 (91.5) 2,609 (92.5)
Female 1,625 (57.6) 93–95 225 (8.0) 188 (6.7)
Male 1,197 (42.4) #92 16 (0.6) 25 (0.9)
Age (years) Notes: “Pack-years” were obtained by multiplying the average number of cigarettes
,70 2,162 (76.6) 1,346 (47.7) smoked daily by the number of years smoked and dividing the product by 20. adyspnea:
$70 660 (23.4) 1,476 (53.3) 0= no dyspnea, 1= dyspnea walking rapidly on level ground or up a moderate slope,
and $2= dyspnea walking calmly on level ground, washing or dressing, or at rest;
Self-reported diseases b
upper limits: men 17 g/dL, women 16 g/dL; cupper limits: men 480 μmol/L, women
CVD 348 (12.3) 541 (19.2) 18–49 years 350 μmol/L, women $50 years 400 μmol/L.
 Asthma 218 (7.7) 288 (10.2) Abbreviations: BMI, body mass index; COPD, chronic obstructive pulmonary
COPD 115 (4.1) 178 (6.3) disease; CRP, C-reactive protein; CVD, cardiovascular disease; FEV1, forced
Diabetes 87 (3.1) 198 (7.0) expiratory volume in 1 second; FVC, forced vital capacity; SpO2, arterial oxygen
saturation as measured by pulse oximetry.
 Hypertension 571 (20.2) 1,018 (36.1)
Smoking history
 Never smoked 1,010 (35.8) 1,008 (35.7) Table 3 shows the results of the multivariable logistic
Former smoker 1,143 (40.5) 1,386 (49.1)
630 (22.3) 816 (28.9)
regression. In the final multivariable analysis, 2,682 subjects
$10 pack-years
Current smoker 669 (23.7) 428 (15.2) were included. Current smoking with pack-years $10 and
$10 pack-years 558 (19.8) 387 (13.7) FEV1 % predicted ,50 had the highest odds ratio (OR),
Dyspneaa 2.48 (1.48–4.15) and 3.55 (1.60–7.89), respectively. In addi-
0 1,557 (55.2) 1,335 (47.3)
1 1,120 (39.7) 1,266 (44.9) tion age, male sex, former smoking with pack-years $10,
$2 145 (5.1) 221 (7.8) increased CRP, and high BMI were significant predictors in
Chronic cough with sputum the multivariable analysis. We did not find any significant
 No 2,646 (93.8) 2,637 (93.4)
interactions. Assumptions for logistic regression were met,
Yes 176 (6.2) 185 (6.6)
BMI (kg/m2) and we did not find any multicollinearity.
,20 57 (2.0) 75 (2.7) When FEV1/FVC ratio, as a dichotomous (with a thresh-
20–30 2,171 (76.9) 2,163 (76.7)
old of 0.7 or lower) or as a continuous variable, was added to
$30 585 (20.8) 583 (20.7)
FEV1 % predicted the multivariable model that included FEV1 % predicted, no
$80 2,018 (71.5) 2,089 (74.0) significant association with SpO2 decrease was found; 91.7%
50–80 708 (25.1) 614 (21.8) of FEV1 % predicted ,50 had FEV1/FVC ratio ,0.7.
,50 48 (1.7) 65 (2.3)
Frequency of SpO2 decrease by changes from baseline
FEV1/FVC ratio
,70 631 (23.1) 805 (29.5) of other variables (across Tromsø 5 and Tromsø 6) is shown
$70 2,097 (76.9) 1,923 (70.5) in Table 4. Participants who had been diagnosed with
Spirometry pattern asthma, COPD, or diabetes between the two time points
 Normal 1,923 (70.5) 1,775 (65.1)
Obstructive 631 (23.1) 805 (29.5)
had a significantly higher incidence of SpO2 decrease.
 Restrictive 174 (6.4) 148 (5.4) These participants had significantly decreased FEV1 %
CRP (mg/L) predicted at baseline, of 81.1, 71.5, and 82.8, respectively,
,5 2,426 (86.0) 2,438 (86.4)
(P,0.001). In addition, FEV1 % predicted decrease/year
$5 357 (12.7) 340 (12.0)
Hemoglobin (g/dL)b and CRP increase were associated with SpO2 decrease. As
#Upper limits 2,546 (90.2) 2,741 (97.1) BMI increased, the frequency of decline in SpO2 increased.
.Upper limits 8 (0.3) 37 (1.3) The opposite was the case when BMI dropped. This trend
Fibrinogen (g/L)
was not statistically significant. Smoking cessation between
,4 2,304 (81.6) 1,943 (68.9)
$4 312 (11.1) 834 (29.6) the time points was associated with a lower frequency of
Uric acid (μmol/L)c SpO2 decrease than was continued smoking, but this finding
#Upper limits 2,547 (90.3) 2,601 (92.2) was not statistically significant.
.Upper limits 110 (3.9) 177 (6.3)
Albumin:creatinine ratio (mg/mmol)
,3 2,606 (92.3) 2,475 (87.7) Discussion
$3 127 (4.5) 231 (8.2) SpO2 decrease was associated with smoking history $10
(Continued) pack-years, lung function (by FEV1 % predicted ,50),

International Journal of COPD 2014:9 submit your manuscript | www.dovepress.com

1229
Dovepress
Vold et al Dovepress

Table 2 Frequency of SpO2 decrease by subjects’ characteristics Table 2 (Continued)

among 2,822 study participants SpO2 (%) P-valuea
SpO2 (%) P-valuea decrease (n)
decrease (n) Uric acid (μmol/L)e
Total numbers 139 (4.9) #Upper limits 122 (4.8) 0.014
Sex .Upper limits 11 (10.3)
Female 59 (3.6) ,0.001 Albumin:creatinine ratio (mg/mmol)
Male 80 (6.7) ,3 127 (4.9) 0.06
Age (years) $3 11 (8.7)
,70 96 (4.4) 0.031 Notes: “Pack-years” were obtained by multiplying the average number of cigarettes
$70 43 (6.5) smoked daily by the number of years smoked and dividing the product by 20. aP-values
Self-reported diseases by chi-square; for smoking compared with never smoking, for others by trend; bdyspnea:
0= no dyspnea, 1= dyspnea walking rapidly on level ground or up a moderate slope, and
CVD 27 (7.8) 0.009 $2= dyspnea walking calmly on level ground, washing or dressing, or at rest; cupper
 Asthma 12 (5.5) 0.7 limits men 17 g/dL, women 16 g/dL; dFischer’s exact test; eupper limits men 480 μmol/L,
COPD 9 (7.8) 0.1 women 18–49 years 350 μmol/L, women $50 years 400 μmol/L.
Diabetes 7 (8.0) 0.2 Abbreviations: BMI, body mass index; CRP, C-reactive protein; COPD, chronic
obstructive pulmonary disease; CVD, cardiovascular disease; FEV1, forced expiratory
 Hypertension 28 (4.9) 1.0
volume in 1 second; FVC, forced vital capacity; SpO2, arterial oxygen saturation as
Smoking history measured by pulse oximetry.
 Never smoking 30 (3.0)
Former smoking 65 (5.7) 0.002
,10 pack-years 21 (4.1) 0.2 BMI $30 kg/m2, CRP $5 mg/L, male sex, and age, in multi-
$10 pack-years 44 (7.0) ,0.001 variable logistic regression. Decrease in FEV1 % predicted was
 Smoking 44 (6.6) ,0.001 significantly associated with decrease in SpO2, and a tendency,
,10 pack-years 2 (1.8) 0.8
although not statistically significant, for less frequent SpO2
$10 pack-years 42 (7.5) ,0.001
Dyspneab decrease with BMI decrease and smoking cessation were also
0 68 (4.4) 0.3 observed.
1 62 (5.5)
$2 9 (6.2)
Chronic cough with sputum Table 3 Factors associated with arterial oxygen saturation (SpO2)
 No 122 (4.6) 0.003 decrease in multivariable logistic regression
Yes 17 (9.7)
OR 95% CI P-value
BMI (kg/m2)
,20 3 (5.3) 0.033 Sex
20–30 95 (4.4) Male 1.68 1.15–2.45 0.008
$30 41 (7.0) Age (years) 1.03 1.01–1.06 0.008
FEV1 % predicted Smoking history
$80 84 (4.2) ,0.001  Never smoking 1
50–80 44 (6.2) Former smoking
,50 9 (18.8) ,10 pack-years 1.14 0.62–2.09 0.7
FEV1/FVC ratio $10 pack-years 1.74 1.04–2.92 0.035
,70 39 (6.2) 0.08  Smoking
$70 94 (4.5) ,10 pack-years 0.71 0.17–3.04 0.6
Spirometry pattern $10 pack-years 2.48 1.48–4.15 0.001
 Normal 78 (4.1) 0.001 BMI (kg/m2)
Obstructive 39 (6.2) ,30 1
 Restrictive 16 (9.2) $30 1.72 1.15–2.57 0.008
CRP (mg/L) FEV1 % predicted
,5 107 (4.4) ,0.001 $80 1
$5 32 (9.0) 50–80 1.18 0.79–1.75 0.4
Hemoglobin (g/dL)c ,50 3.55 1.60–7.89 0.002
#Upper limits 126 (4.9) 0.7d CRP (mg/L)
.Upper limits 0 (0.0) $5 1.74 1.12–2.71 0.013
Fibrinogen (g/L) Notes: N=2,682 subjects. Pack-years were obtained by multiplying the average
,4 107 (4.6) 0.011 number of cigarettes smoked daily by the number of years smoked and dividing the
25 (8.0) product by 20.
$4
Abbreviations: BMI, body mass index; CI, confidence interval; CRP, C-reactive
(Continued) protein; FEV1, forced expiratory volume in 1 second; OR, odds ratio.

1230 submit your manuscript | www.dovepress.com International Journal of COPD 2014:9

Dovepress
Dovepress Factors associated with oxygen saturation decrease

Table 4 Frequency of SpO2 decrease and associations to changes We found a signif icant association between SpO 2
from baseline characteristics decrease and self-reported CVD in univariable analysis.
Total SpO2 (%) P-value CVD contributes to heart failure, which may affect pulmo-
(n) decrease (n)
nary function and thus lower SpO2. This association was
2,822 139 (4.9)
not shown in multivariable analysis. One reason for this
Self-reported diseases, new
CVD 225 7 (3.1) 0.2 might be that CVD is strongly associated with both age
 Asthma 104 14 (13.5) ,0.001 and male sex.
COPD 117 16 (13.7) ,0.001 Low SpO2 and partial pressure of oxygen in arterial
Diabetes 113 10 (8.8) 0.049
blood (PaO2) in smokers have been shown in previous
 Hypertension 512 28 (5.5) 0.5
Smoking history T5–T6 0.3 studies. 28,29 Even when correcting for lung function
Quit smoking 266 14 (5.3) by FEV 1 % predicted, this association was clearly
Continued smoking 399 30 (7.5) demonstrated.
BMI (kg/m2)
More than 90% of the group with a FEV1 % predicted
 All 0.09
$2↑ 302 22 (7.3) 50 had an FEV1/FVC ratio 0.7. Even though an FEV1/
2↑–2↓ 2,201 106 (4.8) FVC ratio 0.7 was not significant in univariate analysis,
$2↓ 309 11 (3.6) severe airflow limitation seems to be associated with an
$30 0.1
SpO2 decrease.
$2↑ 60 7 (11.5)
407 29 (7.1)
We found that baseline CRP $5 mg/L was associ-
2↑–2↓
$2↓ 115 4 (3.5) ated with an SpO2 decrease in both uni- and multivari-
FEV1 % predicted/yeara able analysis, and the associations with CVD and other
$2↓ 168 18 (10.7) ,0.001 chronic diseases probably contributed to increased OR
CRP (mg/L)
in the multivariable analysis. Other biomarkers, such as
$5↑ 142 13 (9.2) 0.020
Fibrinogen (g/L) fibrinogen, uric acid, and microalbuminuria (expressed
$1↑ 593 33 (5.6) 0.6 by the albumin:creatinine ratio), were significant in uni-
Uric acid (μmol/L) variable, but not multivariable, analyses. Microalbuminuria
$60↑ 293 15 (5.1) 0.9 has been found to be associated with hypoxia (defined as
Albumin:creatinine ratio (mg/mmol)
SpO2 #92%) in COPD.30,31 In our study, less than 1% of
$3↑ 142 8 (5.6) 0.7
Notes: Decrease divided by years between examinations. Upward arrows indicate
a
participants had SpO2 #92%, which may be a reason for
an increase, downward arrows indicate a decrease. not finding this association. CRP might also be a better
Abbreviations: BMI, body mass index; COPD, chronic obstructive pulmonary
disease; CRP, C-reactive protein; CVD, cardiovascular disease; FEV1, forced
marker of inflammation associated with SpO2 decrease than
expiratory volume in 1 second; SpO2, arterial oxygen saturation as measured by microalbuminuria, fibrinogen, and uric acid.
pulse oximetry; T5, Tromsø 5; T6, Tromsø 6.
BMI $30 kg/m2 was, as expected, associated with SpO2
decrease. Among other disorders, this group is at risk for
Comparison with previous studies sleep apnea and obesity hypoventilation, which is known to
This study complements our previous cross-sectional study show- lead to low daytime SpO2.32–34
ing that smoking, FEV1 % predicted, and obesity are the most Baseline hemoglobin above the upper limit was not asso-
important predictors of low SpO2.10 The findings from a longitu- ciated with SpO2 decrease. This was expected since a high
dinal cohort provide stronger indications of a causal relationship hemoglobin value is usually a consequence of, rather than a
than can be determined using a cross-sectional study.27 reason for, a decrease in SpO2.
The association between male sex and a fall in SpO2 was A new diagnosis of asthma or COPD between the time
consistent with our previous findings. More men had previ- points was associated with a decrease in SpO2. Subjects with
ously smoked and had also smoked for more pack-years. such a new diagnosis had decreased FEV1 % predicted at
CVD was also more common in men. baseline. COPD is frequently underdiagnosed, which may
The impact of age was also consistent with the cross- be linked to less help-seeking among smokers.35 It is not
sectional study. Aging means physiological changes and unexpected that subjects recently diagnosed with COPD
increasing comorbidity, and the summation of risk factors or asthma had troubling symptoms and increased risk of
might accelerate an SpO2 decrease. decreased SpO2.

International Journal of COPD 2014:9 submit your manuscript | www.dovepress.com

1231
Dovepress
Vold et al Dovepress

Strength and limitation the reference values, when applied in the oldest age groups,
The subjects in this study were a subgroup of participants might also have played a role.
in the cross-sectional study on SpO2 from the sixth Tromsø
Study.10 This study would have provided stronger supple- Clinical implications
mental evidence if the subjects had been recruited from a This study describes associations between unhealthy lifestyle
separate population sample. Of the original group examined and decreased SpO2. Smoking stands out as an important
with lung function tests in Tromsø 5, only 54.8% were cause, and not only through its deteriorating effect on
reexamined in Tromsø 6. We know that almost 10% died lung function. Obesity is another modifiable risk factor for
between these time points. Those with severe health prob- decreased SpO2. It is promising that the findings in this study
lems and increased risk of low SpO2 probably participated indicate that subjects who stop smoking or lose weight may
to a lesser degree than others. We found that almost 10% have a decreased risk of decreased SpO2. It may be possible
quit smoking, mean FEV1 % predicted increased, and those to stabilize SpO2 with a healthier lifestyle.
in the obese category lost weight. A healthy survivor effect
and a decreased representation of those with poor health may Conclusion
have led to a healthier sample. This may explain why aging A decrease in SpO2 was most strongly associated with low
did not lead to decreased SpO2. FEV1 % predicted and a history of smoking. It was also
Smoking may have been a difficult topic for some partici- associated with higher BMI. This is in accordance with the
pants, and thus there may have been some bias in categorizing findings of our previous cross-sectional study. Smoking ces-
smokers, former smokers, and never smokers. Yet previous sation and reducing obesity are important measures that may
studies have showed that self-reports of smoking are usually help avoid SpO2 decrease in the general population.
accurate.36,37 The pack-years calculated might be uncertain,
because of recall bias, especially among former smokers. Author contributions
Only seven out of 256 participants who quit smoking between All authors participated in concept and design of the study.
Tromsø 5 and 6 had valid data on the question, “How long HM performed data collection, and MLV and HM performed
has it been since you stopped?” Some participants may have data analysis and interpretation. MLV and HM drafted the
stopped smoking recently, and the effect of smoking cessa- manuscript. All authors participated in revision and gave final
tion on SpO2 may not have been measurable yet, thereby approval of the manuscript.
weakening the associations.
Pulse oximetry has some limitations; among others, high Disclosure
carboxyhemoglobin might have given falsely elevated SpO2 The authors report no conflict of interest in this work.
in smokers and thus, diminished the association between
SpO2 and smoking.38 References
SpO2 decrease may be imprecise since the accuracy of 1. Kent BD, Mitchell PD, McNicholas WT. Hypoxemia in patients with
COPD: cause, effects, and disease progression. Int J Chron Obstruct
the device is within ±2%. By using three measurements, Pulmon Dis. 2011;6:199–208.
categorizing in groups, and excluding values that fell within 2. West JB. Causes of and compensations for hypoxemia and hypercapnia.
Compr Physiol. 2011;1(3):1541–1553.
the normal range (SpO2 .95%), we tried to decrease this 3. Young IH, Bye PT. Gas exchange in disease: asthma, chronic obstructive
influence. pulmonary disease, cystic fibrosis, and interstitial lung disease. Compr
The FEV1 % predicted values in Tromsø 6 increased. Physiol. 2011;1(2):663–697.
4. Crapo RO, Jensen RL, Hegewald MATH, Taskin DP. Arterial blood gas
­Reasons for this might be selection bias and a healthier reference values for sea level and an altitude of 1,400 meters. Am J Respir
sample. Using two different spirometers may also have Crit Care Med. 1999;160(5):1525–1531.
5. Little SA, Elkholy MM, Chalmers GW, Farouk A, Patel KR, Thomson NC.
contributed to the difference, even though we attempted to Predictors of nocturnal oxygen desaturation in patients with COPD.
correct for this. Using age in whole years as of December 31 Respir Med. 1999;93(3):202–207.
in calculating FEV1 % predicted may have resulted in a sys- 6. Knower MT, Dunagan DP, Adair NE, Chin R Jr. Baseline oxygen
saturation predicts exercise desaturation below prescription threshold in
tematic bias, since the mean years between analyses were patients with chronic obstructive pulmonary disease. Arch Intern Med.
6.3, not 6.0;39 however, this is likely to have underestimated 2001;161(5):732–736.
7. Edvardsen A, Akero A, Christensen CC, Ryg M, Skjonsberg OH. Air
the increase in predicted values. Due to the high number of travel and chronic obstructive pulmonary disease: a new algorithm for
subjects aged 75 years or older, a limitation in the validity of pre-flight evaluation. Thorax. 2012;67(11):964–969.

1232 submit your manuscript | www.dovepress.com International Journal of COPD 2014:9

Dovepress
Dovepress Factors associated with oxygen saturation decrease

8. Canet J, Gallart L, Gomar C, et al. Prediction of postoperative pulmonary 24. Kohansal R, Martinez-Camblor P, Agusti A, Buist AS, Manninio DM,
complications in a population-based surgical cohort. Anesthesiology. Soriano JB. The natural history of chronic airflow obstruction revisited:
2010;113(6):1338–1350. an analysis of the Framingham offspring cohort. Am J Respir Crit Care
9. Roberts CM, Bugler JR, Melchor R, Hetzel MR, Spiro SG. Value of Med. 2009;180(1):3–10.
pulse oximetry in screening for long-term oxygen therapy requirement. 25. Fletcher C, Peto R. The natural history of chronic airflow obstruction.
Eur Respir J. 1993;6(4):559–562. Br Med J. 1977;1(6077):1645–1648.
10. Vold ML, Aasebø U, Hjalmarsen A, Melbye H. Predictors of oxygen 26. Kunzli N, Kuna-Dibbert B, Keidel D, et al. Longitudinal validity of
saturation #95% in a cross-sectional population based survey. Respir spirometers – a challenge in longitudinal studies. Swiss Med Wkly.
Med. 2012;106(11):1551–1558. 2005;135(33–34):503–508.
11. Ahmadi-Abhari S, Kaptoge S, Luben RN, Wareham NJ, Khaw KT. 27. Szklo M, Nieto FJ. Epidemiology: Beyond the Basics. 2nd ed.
Longitudinal association of C-reactive protein and lung function over ­Massachusetts: Jones and Bartlett Publishers; 2007.
13 years: The EPIC-Norfolk study. Am J Epidemiol. 2014;179(1): 28. Tirlapur VG, Gicheru K, Charalambous BM, Evans PJ, Mir MA. Packed
48–56. cell volume, haemoglobin, and oxygen saturation changes in healthy
12. Kalhan R, Tran BT, Colangelo LA, et al. Systemic inflammation in smokers and non-smokers. Thorax. 1983;38(10):785–787.
young adults is associated with abnormal lung function in middle age. 29. Ouattara S, Keita M, Tuo N, Dah C, Siransy EA, Bogui P. [Effect of
PLoS One. 2010;5(7):e11431. smoking on PaO2 at rest and during moderate exercise]. Dakar Med.
13. Rosenberg SR, Kalhan R. Biomarkers in chronic obstructive pulmonary 2002;47(1):90–95.
disease. Transl Res. 2012;159(4):228–237. 30. Casanova C, de Torres JP, Navarro J, et al. Microalbuminuria and
14. Agusti A, Soriano JB. COPD as a systemic disease. COPD. hypoxemia in patients with chronic obstructive pulmonary disease. Am
2008;5(2):133–138. J Respir Crit Care Med. 2010;182(8):1004–1010.
15. Vestbo J, Hurd SS, Agusti AG, et al. Global strategy for the ­diagnosis, 31. Komurcuoglu A, Kalenci S, Kalenci D, Komurcuoglu B, Tibet G.
management, and prevention of chronic obstructive pulmonary Microalbuminuria in chronic obstructive pulmonary disease. Monaldi
disease: GOLD executive summary. Am J Respir Crit Care Med. Arch Chest Dis. 2003;59(4):269–272.
2013;187(4):347–365. 32. Laaban JP, Cassuto D, Orvoen-Frija E, et al. Cardiorespiratory conse-
16. Kapsimalis F, Varouchakis G, Manousaki A, et al. Association of sleep quences of sleep apnoea syndrome in patients with massive obesity.
apnea severity and obesity with insulin resistance, C-reactive protein, Eur Respir J. 1998;11(1):20–27.
and leptin levels in male patients with obstructive sleep apnea. Lung. 33. Verin E, Tardif C, Pasquis P. Prevalence of daytime hypercapnia or
2008;186(4):209–217. hypoxia in patients with OSAS and normal lung function. Respir Med.
17. Roche F, Gaspoz JM, Pichot V, et al. Association between C-reactive 2001;95(8):693–696.
protein and unrecognised sleep-disordered breathing in the elderly. Eur 34. Zavorsky GS, Hoffman SL. Pulmonary gas exchange in the morbidly
Respir J. 2009;33(4):797–803. obese. Obes Rev. 2008;9(4):326–339.
18. Vijayan VK. Morbidities associated with obstructive sleep apnea. Expert 35. Melbye H, Joensen L, Risor MB, Halvorsen PA. Symptoms of respira-
Rev Respir Med. 2012;6(5):557–566. tory tract infection and associated care-seeking in subjects with and
19. Aksu F, Capan N, Aksu K, et al. C-reactive protein levels are raised without obstructive lung disease; the Tromso Study: Tromso 6. BMC
in stable chronic obstructive pulmonary disease patients indepen- Pulm Med. 2012;12:51.
dent of smoking behavior and biomass exposure. J Thorac Dis. 36. Patrick DL, Cheadle A, Thompson DC, Diehr P, Koepsell T, Kinne S.
2013;5(4):414–421. The validity of self-reported smoking: a review and meta-analysis. Am
20. de Torres JP, Cordoba-Lanus E, Lopez-Aguilar C, et al. C-reactive J Public Health. 1994;84(7):1086–1093.
protein levels and clinically important predictive outcomes in stable 37. Vartiainen E, Seppala T, Lillsunde P, Puska P. Validation of self
COPD patients. Eur Respir J. 2006;27(5):902–907. reported smoking by serum cotinine measurement in a community-based
21. Jacobsen BK, Eggen AE, Mathiesen EB, Wilsgaard T, Njolstad I. Cohort study. J Epidemiol Community Health. 2002;56(3):167–170.
profile: the Tromso Study. Int J Epidemiol. 2011;41(4):961–967. 38. Jensen LA, Onyskiw JE, Prasad NG. Meta-analysis of arterial oxy-
22. Miller MR, Hankinson J, Brusasco V, et al. Standardisation of spirom- gen saturation monitoring by pulse oximetry in adults. Heart Lung.
etry. Eur Respir J. 2005;26(2):319–338. 1998;27(6):387–408.
23. Langhammer A, Johnsen R, Gulsvik A, Holmen TL, Bjermer L. Forced 39. Quanjer PH, Hall GL, Stanojevic S, Cole TJ, Stocks J. Age- and height-
spirometry reference values for Norwegian adults: the Bronchial Obstruc- based prediction bias in spirometry reference equations. Eur Respir J.
tion in Nord-Trondelag Study. Eur Respir J. 2001;18(5):770–779. 2012;40(1):190–197.

International Journal of COPD Dovepress

Publish your work in this journal
The International Journal of COPD is an international, peer-reviewed This journal is indexed on PubMed Central, MedLine and CAS. The
journal of therapeutics and pharmacology focusing on concise rapid manuscript management system is completely online and includes a
reporting of clinical studies and reviews in COPD. Special focus is given very quick and fair peer-review system, which is all easy to use. Visit
to the pathophysiological processes underlying the disease, intervention http://www.dovepress.com/testimonials.php to read real quotes from
programs, patient focused education, and self management protocols. published authors.
Submit your manuscript here: http://www.dovepress.com/international-journal-of-copd-journal

International Journal of COPD 2014:9 submit your manuscript | www.dovepress.com

1233
Dovepress