Rheumatology 2008;47;219221

doi:10.1093/rheumatology/kem353

Concise Report

Sparing of the thumb in Raynauds phenomenon

B. Chikura1, T. L. Moore2, J. B. Manning2, A. Vail3 and A. L. Herrick2

KEY

WORDS:

Raynauds phenomenon, Thumb, Thermography, Systemic sclerosis.

was 46 yrs and the range was 1871 yrs. The median duration of
symptoms was 47 months, range 2540 months. Only patients
with definite RP (defined by at least two colour changes, one of
them being white in response to cold or emotion), 18 yrs or older
and able to give consent were recruited.
Fourteen patients (2 males, 12 females) fulfilled the LeRoy and
Medsger [6] criteria for PRP. Thirty patients (4 males, 26 females)
were classified as SRP [6, 7]. Twelve patients had SSc, 10 patients
had the limited cutaneous and two had the diffuse cutaneous
subtype [8]. Thirteen patients had undifferentiated CTD (UCTD)
(2 males, 11 females). These patients had features that precluded
the diagnosis of PRP but did not have a specific CTD. All patients
with UCTD were antinuclear antibody (ANA)-positive (positivity
was defined by an IgG titre of >1/100). Nailfold capillaroscopy
was done in 9 of these 13 patients and three had abnormal
findings. One patient had mixed CTD (MCTD) and was positive
for UI-RNP antibody, three patients (all female) had primary
Sjogrens syndrome (one patient was positive for anti-Ro/La
antibodies) and one patient had cold agglutinin disease.
Twelve patients (27%) were on vasoactive drugs such as
calcium channel blockers and angiotensin-converting enzyme
inhibitors. Of the 12 patients who were on vasoactive drugs, 10
patients (83%) fulfilled the criteria for SRP, and two patients
(17%) the criteria for PRP. Twenty-four patients (55%) had a
history of smoking and 13 patients (30%) were current smokers.

Introduction
In RP, classically the digits turn white (vasospasm), then blue
(deoxygenation of static venous blood), then red (reperfusion) in
response to precipitating factors such as cold exposure or
emotional stress [1, 2]. RP can be divided into primary and
secondary depending on absence or presence of an underlying
disease/disorder, respectively. For example, connective tissue
diseases (CTDs) such as SSc [3].
There are many methods that have been developed over the
years to quantify digital vascular involvement in RP. These
include nailfold capillaroscopy, laser Doppler flowmetry, thermography and finger systolic blood pressure [4, 5]. It has been
observed using these techniques that digital involvement is not
uniform and anecdotally, many patients when specifically asked
report that the thumb is spared. Yet, to our knowledge there are
no previous studies that have looked into thumb involvement. If
the thumb is not involved, this raises the question as to why this is
the case. The other question is whether thumb involvement is of
prognostic value, i.e. do those patients who have RP affecting
thumbs have more severe disease than those without? The
difference in the degree of thumb involvement between primary
Raynauds phenomenon (PRP) and secondary Raynauds phenomenon (SRP) has not been previously investigated. This could
be clinically important and might assist in clinical classification. If
there is a difference in thumb involvement between PRP and SRP
this could imply different pathophysiologies.

Methods

Patients and methods

Symptoms. The following information was recorded by

patients to assess the frequency and severity of symptoms of RP:

Patients
Forty-four patients (6 males, 38 females) were recruited at Hope
Hospital between November 2006 and April 2007. The median age

(i) Colour changes in digits during an attack of RP (options of

white, blue and red were given to patients and they could
select one or more of the three colour changes) as noted by
the patient.
(ii) The frequency of symptoms was graded on a scale of never,
sometimes and always involved for each digit as noted by
the patient.
(iii) The worse of the left and right scores for each digit (never
being the best of the three scores, always being the worst
and sometimes in between) were considered for analysis.
For example, if the left thumb score was sometimes and the

1
The Royal Liverpool University Hospital, Liverpool, 2Rheumatic Diseases
Centre and 3Biostatistics Group, University of Manchester, Hope Hospital,
Salford, UK.

Submitted 21 September 2007; revised version accepted 30 November 2007.

Correspondence to: B. Chikura, The Royal Liverpool University Hospital,
Prescot Street, Liverpool L7 8XP, UK. E-mail: docbatsi@aol.com

219
The Author 2008. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Downloaded from http://rheumatology.oxfordjournals.org/ by guest on December 24, 2014

Objectives. To conduct a prospective study to determine which digits are affected (and whether the thumb is spared or not) in a cohort of
patients with RP as assessed by symptoms and thermography and to determine whether the degree of thumb involvement differs between
primary (PRP) and secondary Raynauds phenomenon (SRP).
Methods. This was a cross-sectional study of 44 patients with RP. The following characteristics were recorded to allow comparisons between
digits: symptoms of RP in each digit (graded on a scale of never, sometimes and always affected during an attack of RP) and
thermography at 238C. A distaldorsal difference (DDD) in temperature at 238C of
18C or less was considered to be clinically relevant.
Results. Symptom scores in the thumb were significantly better, i.e. less severe than in each finger (P < 0.001). As only three participants
had any finger better than the thumb, there was no power to compare whether the thumb was spared more in PRP compared with SRP.
Mean DDD was significantly higher (i.e. better) in the thumb compared with each finger (P < 0.001). Although DDD scores were higher in PRP
compared with SRP (P 0.01), there was no evidence that the relative effect of the thumb differed between the two groups (P 0.26).
Conclusions. Our findings confirm that the thumbs are spared in RP, both primary and secondary, as demonstrated by both symptoms and
thermography. The reasons for sparing of the thumb were not addressed in this study but raised questions regarding pathophysiolgy.

B. Chikura et al.

220

right thumb was never, sometimes involved was considered for analysis.
The study was approved by the Warrington, Wigan and Leigh
Research Ethics Committee and all subjects gave informed
consent.

Sample size. It was calculated that 16 patients per group (i.e.

PRP and SRP) would allow detection of a difference of 1 S.D. in
the outcome (e.g. the average temperature gradient along the
thumbs) with 80% power. However, it was considered that the
ratio between PRP and SRP would be 1:3, requiring 11 patients
with PRP and 33 with SRP.

Never

Always

Total, n (%)

Thumb vs index finger

Never
1
Sometimes
6
Always
11
Total, n (%) 18 (41)

0
7
15
22 (50)

0
0
4
4 (9)

1 (2)
13 (30)
30 (68)
44

<0.001

Thumb vs middle finger

Never
0
Sometimes
4
Always
14
Total, n (%) 18 (41)

0
9
13
22 (50)

0
0
4
4 (9)

0 (0)
13 (30)
31 (70)
44

<0.001

Thumb vs ring finger

Never
1
Sometimes
6
Always
11
Total, n (%) 18 (41%)

0
11
11
22 (50)

0
2
2
4 (9)

1 (2)
19 (43)
24 (55)
44

<0.001

Thumb vs little finger

Never
2
Sometimes
8
Always
8
Total, n (%) 18 (41)

1
13
8
22 (50)

0
2
2
4 (9)

3 (7)
23 (52)
18 (41)
44

<0.001

This table shows comparison of reporting of symptoms of always, sometimes and never
between the thumbs and other digits. The worst score from each pair of digits was considered for
analysis. The percentages of the totals for symptom frequencies are shown to allow comparison
between digits. The P-values obtained using the McNemars tests for each comparison are
shown.

TABLE 2. DDD in the different digits showing (i) numbers of digits with a clinically
relevant temperature gradient, i.e. DDD of
18C or less and (ii) the mean DDD for
each digit for the whole patient cohort and for the subgroups with PRP and SRP.

Statistical analysis. The results were analysed using

McNemars test and repeated measures analyses of variance
(ANOVA).
Results
Symptoms
Comparison of symptom frequencies on a scale of always,
sometimes and never did not show that digits were different
with the exception of the thumb, which was less frequently
involved (P < 0.001) (Table 1). Eighteen patients (41%) had never
had their thumbs affected by RP as compared with none for the
middle finger, i.e. all patients reported middle finger involvement.
Four patients (9%) indicated that their thumbs were always
involved, and in comparison 31 patients (70%) reported that their
middle fingers were always involved. As only three participants
had any finger better than the thumb, there was no power to
compare PRP and SRP.

Thermography
Proportions of different digits with a DDD of
18C or
less. The proportions of thumbs with clinically relevant DDD,
i.e. a DDD of
18C or less, showed that the thumb was the least
affected digit; 18% of thumbs compared with 64% in the index
finger had clinically relevant DDD (Table 2).

Absolute temperature gradients of digits. The thumbs were

warmer compared with other digits (Table 2). The mean DDD for
the thumbs was significantly higher, i.e. warmer, compared with
other digits (P < 0.001). Patients with PRP had higher values than
those with SRP (P 0.01, ANOVA) (Table 2). Although there
were greater differences between thumb and other finger DDD
scores in SRP compared with PRP, these were not statistically
significant (P 0.26) and this could be due to low power to detect
the interaction.

McNemars
P-value

Sometimes

DDD of
18C
or less at 238C
(All patients), n (%)

Thumb
Index
Middle
Ring
Little

8
28
24
25
23

(18)
(64)
(55)
(57)
(52)

Mean DDD at 238C, 8C

All patients
(N 44)

PRP
(N 14)

SRP
(N 30)

0.09

0.82

1.02

0.98

1.19

0.63
0.52
0.11

0.03

0.49

0.43

1.44

1.54

1.42

1.50

ANOVA
P-value
(vs thumb)

<0.001
<0.001
<0.001
<0.001

DDD, distal dorsal difference. Values are the worse score from each pair of digits.

Discussion
This is the first study to look specifically at thumb sparing in
RP. Our findings show that the thumb is spared as reported
by patients. Symptoms were comparable in all digits with
the exception of the thumb, which was significantly less frequently
involved. This was confirmed by thermography. The mean
DDD of the thumbs was higher (i.e. their tips were warmer)
compared with other digits, and the proportion of thumbs with
a DDD of
18C or less at 238C was lower than that for the
other digits.
The degree of thumb sparing was more pronounced in the SRP
group compared with PRP group (Table 2). Although DDD
scores were higher (better) in PRP compared with SRP, there was
no evidence that the relative effect of the thumb differed between
the PRP and SRP. If the thumb is more involved in SRP as
compared with PRP, this observation raises the question as to
whether thumb involvement is of prognostic value, i.e. do those
patients who have RP affecting the thumb have more severe
disease than those without? The question as to whether thumb
involvement differs between PRP and SRP is clinically relevant to
a rheumatologist in that this may help distinguish the two
conditions and imply different pathophysiologies. Vascular
problems in PRP are mainly functional; in contrast, structural
vascular abnormalities can be seen in patients with SRP,

Downloaded from http://rheumatology.oxfordjournals.org/ by guest on December 24, 2014

Thermography. The camera used was an Agema 570 infrared

thermography camera (Flir Systems Limited) with the Agema
Research 2.1 programme software installed on a standard desktop
computer. Thermography was performed by a senior vascular
technician using a standard thermography protocol used at Hope
Hospital. Patients were asked to refrain from caffeine/nicotine for
4 h prior to testing. After acclimatization at 238C for 20 min in a
temperature-controlled room an image of the dorsum of each
hand was taken, and from these hand images the distaldorsal
difference (DDD) was calculated, i.e. the temperature of the tip of
the fingers minus the temperature of the dorsum of the hand.
A DDD of
18C or less at 238C was considered to be clinically
relevant involvement [9, 10]. The worst score (the lower score, i.e.
the more negative value) from each pair of digits was considered
for analysis. Results of DDD at 238C were compared between the
fingers and the thumbs to assess the degree of thumb involvement
and to compare between PRP and SRP.

TABLE 1. Comparison of frequency of symptoms between thumbs and other digits

Sparing of the thumb in Raynauds phenomenon

thumb were not addressed in this study but raise questions

regarding pathophysiology. Thumb involvement may have prognostic value, pointing to an underlying CTD and/or increased
severity of RP. This hypothesis should be tested in prospective
studies.

Rheumatology key messages

 The thumb is spared in both PRP and SRP, as evidenced by both
symptoms and thermography.
 Sparing of the thumb raises questions about pathophysiology.

Disclosure statement: The authors have declared no conflicts of

interest.

References
1 Raynaud M. Local asphyxia and symmetrical gangrene of the extremities
1862. New researches on the nature and treatment of local asphyxia of the
extremities 1874. Translated by Barlow in 1888. London: New Sydenham Society,
1888:1199.
2 Cooke JP, Marshall JM. Mechanisms of Raynauds disease. Vasc Med
2005;10:293307.
3 Herrick AL. Pathogenesis of Raynauds phenomenon. Rheumatology
2005;44:58796.
4 Herrick AL, Clark S. Quantifying digital vascular disease in patients with
primary Raynauds phenomenon and systemic sclerosis. Ann Rheum Dis
1998;57:708.
5 Del Bianco E, Magini B, Muscarella G, Cappugi P, Lotti T. Raynauds phenomenon
(primary or secondary to systemic sclerosis). The usefulness of laser-Doppler
flowmetry in the diagnosis. Int Angiol 2001;20:30713.
6 LeRoy E, Medsger TJ. Raynauds phenomenon: a proposal for classification.
Clin Exp Rheumatol 1992;10:4858.
7 Kallenberg CG. Early detection of connective tissue disease in patients with
Raynauds phenomenon. Rheum Dis Clin North Am 1990;16:1130.
8 LeRoy EC, Black C, Fleischmajer R et al. Scleroderma (systemic sclerosis):
classification, subsets and pathogenesis. J Rheumatol 1988;15:2025.
9 Anderson ME, Moore TL, Lunt M, Herrick AL. The distal-dorsal difference: a
thermographic parameter by which to differentiate between primary and secondary
Raynauds phenomenon. Rheumatology 2007;46:5338.
10 Clark S, Hollis S, Campbell F, Moore T, Jayson M, Herrick A. The distal-dorsal
difference as a possible predictor of secondary Raynauds phenomenon.
J Rheumatol 1999;26:11258.
11 Black CM, Clark RP, Darton K, Goff MR, Norman TD, Spikes HA. A pyroelectric
thermal imaging system for use in medical diagnosis. J Biomed Eng 1990;12:2816.
12 Caramaschi P, Codella O, Poli G et al. Use of computerized digital thermometry for
diagnosis of Raynauds phenomenon. Angiology 1989;40:86371.
13 Darton K, Black CM. Pyroelectric vidicon thermography and cold challenge quantify
the severity of Raynauds phenomenon. Br J Rheumatol 1991;30:1905.
14 OReilly D, Taylor L, el-Hadidy K, Jayson MI. Measurement of cold challenge
responses in primary Raynauds phenomenon and Raynauds phenomenon
associated with systemic sclerosis. Ann Rheum Dis 1992;51:11936.
15 Coughlin PA, Chetter IC, Kent PJ, Kester RC. The analysis of sensitivity, specificity,
positive predictive value and negative predictive value of cold provocation
thermography in the objective diagnosis of the hand-arm vibration syndrome.
Occup Med 2001;51:7580.
16 von Bierbrauer A, Schilk I, Lucke C, Schmidt JA. Infrared thermography in the
diagnosis of Raynauds phenomenon in vibration-induced white finger. VASA
1998;27:949.
17 Brunelli F, Vigasio A, Valenti P, Brunelli GR. Arterial anatomy and clinical application
of the dorsoulnar flap of the thumb. J Hand Surg 1999;24:80311.
18 De AR, Del MP, Blasetti P, Cervini C. Raynauds phenomenon: clinical spectrum of
118 patients. Clin Rheumatol 2003;22:27984.
19 Brand FN, Larson MG, Kannel WB, McGuirk JM. The occurrence of Raynauds
phenomenon in a general population: the Framingham Study. Vasc Med
1997;2:296301.

Downloaded from http://rheumatology.oxfordjournals.org/ by guest on December 24, 2014

demonstrated by nailfold capillaroscopy. Larger studies are

required to compare thumb involvement between PRP and SRP.
In this study, we have used both subjective (self-reporting of
symptoms of RP) by patients and objective (thermography)
methods to assess the degree of thumb involvement.
Thermography is a non-invasive technique that has been applied
in assessing the degree of digital vascular response to cold [11, 12].
Thermography can help to distinguish between normal, PRP and
SRP [13, 14] and relies on measuring the skin temperature as an
indirect measure of digital blood flow. The DDD in temperature
measures the temperature difference between the fingertips and
dorsum of fingers, i.e. the temperature of finger tips minus the
temperature dorsum of the hands. The value obtained is usually
negative in patients with RP because the dorsum of the hand is
warmer than the tips of fingers. A DDD of less than
18C
(dorsum warmer than the tip of fingers) at 238C can distinguish
between healthy subjects and patients with RP [10]. A DDD of

18C or less was therefore used in this study for clinically relevant
involvement. Thermography has a good positive predictive value
and negative predictive value, in patients with RP, for handarm
vibration syndrome [15, 16] and SSc [9].
This study has generated more questions than answers. We
have demonstrated that the thumb is spared in RP but this raises
the question as to why this is the case. A search for the answer
could lead us to a greater understanding of the pathophysiology of
RP. The anatomical differences between the thumb and other
digits could help explain thumb sparing. It is the shortest and
fattest digit and important for movements that require precision
and a power grip. It is the only digit with two instead of three
phalanges. The blood supply to the thumb differs from that of
other digits. The dorsal arterial blood supply to the thumb is well
developed and is represented by an independent vascular axis, in
contrast to other digits in which the dorsal arteries are inconsistent
and dependent on the palmar circulation [17]. The thumb may be
protected from adverse vascular events such as RP by a rich blood
supply. The length of the digits (and their arteries) could be crucial
in determining the clinical expression of RP. There is some
evidence in this study to support this hypothesis. The thumb is the
shortest digit and is the least involved. The little finger, which is
the second shortest finger, appears to be the second least affected
finger after the thumb, whereas the middle finger, which is the
longest digit, is the worst affected digit as assessed by the
frequency of self-reported symptoms of RP (Table 1). The critical
distance could be the distance between the palmar and dorsal
vascular arches and the tip of the finger as this represents the
length of digital arteries. However, explaining thumb sparing in
RP based solely on anatomical differences may be simplistic.
Our cohort is representative of patients referred to a
rheumatology clinic with RP. Fewer than half of the patients
attending routine rheumatology clinics with RP have a specific
CTD [18]. In our cohort, 16 patients (36%) had a specific CTD. In
secondary care, most RP patients have SRP; however, this
includes patients who have characteristics that preclude the
diagnosis of PRP but have no identifiable CTD. This is in
contrast to general practice in which there are more patients with
PRP (80%) compared with SRP (20%) [19].
In conclusion, our findings confirm that the thumbs are spared
in RP, both primary and secondary, as demonstrated by both
symptoms and thermography. The reasons for sparing of the

221