Prevalence,characteristics and natural history of cold intolerance after the reverse digital artery flap

Introduction

Cold intolerance is characterized by uncomfortable symptoms in previously injured hands felt after exposure to cold weather. Glickmann and Mackinnon (1990) and Morrison et al. (1978) have reported an almost 100% prevalence in a large series of digital replantations. Irwin et al. (1997) found that the prevalence of cold intolerance after nerve injury was 80%. Nijhuis et al. (2010) reported that approximately 38% of 129 hand fracture patients suffered cold intolerance after surgery. Klocker et al. (2012) found that 41% of patients with a vascular injury of the upper limb complained of cold intolerance. Cold intolerance has also been studied in other disorders, such as upper extremity trauma (Koman et al., 1984), Dupuytren’s contracture (King and Belcher, 2014) and Raynaud’s disease (Merla et al., 2002).

Lai et al. (1992) described the reverse digital artery flap and highlighted the advantages of this flap; subsequently, a note of caution was published regarding cold intolerance in the flap, which involved sacrifice of a digital artery (Cormack, 1993). However, few articles have focused on the prevalence, characteristics and natural history of cold intolerance after the flap transfer, especially when used for fingertip and pulp reconstruction.

This study was designed to investigate the prevalence, characteristics and nature history of cold intolerance after a reverse digital artery flap used for fingertip and pulp reconstruction.

Methods

The study group consisted of patients with fingertip amputations or finger pulp defects, who were treated with the reverse digital artery flap in our department from January 2010 to November 2013. In consideration of appearance, the surgical procedure was designed so that the reverse digital artery flap was located at the lateral area on the proximal phalanx of the injured finger (on the ulnar side of the index, middle, ring fingers and on the radial side of the little finger) (Usami et al., 2015).

Study participants were required to meet the following criteria in this retrospective study: single finger soft tissue defect, excluding the thumb; between 16 to 65 years old on the day of injury; and no neurorrhaphy. Patients were excluded when any of the following features were present: previous hand fractures, nerve or vascular injury; previous history of cold intolerance; use of vasodilatation medicines; and combination with another injury. A total of 123 patients fulfilled these criteria. However, 36 of 123 patients were lost to follow-up, leaving 87. The mean follow-up time was 34 months (range 14–61). All these patients had lived through two or more winters after surgery.

The follow-up was conducted by phoning the patients or by face-to-face follow-up. The content of the telephone follow-up included a verbal consent form, the Cold Intolerance Symptom Severity (CISS) questionnaire (Irwin et al., 1997) and a few additional questions. The additional questions included items about smoking history, medication use, hand dominance, the temperature at which patients began to experience cold intolerance, type of injury and any change in symptoms (deteriorated, improved or remained the same).

The first item in the CISS questionnaire was set as an open-ended query and was not scored. The listed symptoms in the first item of the CISS questionnaire (pain, numbness, stiffness, weakness, aching, skin colour change (white/bluish white/blue)) were not disclosed to the patients. We recorded only the first and second symptoms noted by the patients themselves, even if these symptoms were not in the list in the first item of the CISS questionnaire.

The face-to-face follow-up included the same questions and also assessment of sensory recovery in the flap, which was measured by Semmes–Weinstein monofilaments (Aesthesio Precision Sensory Evaluators, DanMic Global, Campbell, CA, USA) and static two-point discrimination with the Disk-Criminator (North Coast Medical, Inc., Gilroy, CA, USA).

We classified the temperatures at which patients began to suffer cold intolerance into four grades: grade 1, <0 °C; grade 2, 0–5 °C; grade 3, 6–10 °C; and grade 4, >10 °C.

Results

A total of 52 out of 87 responders (60%) reported cold intolerance. In the telephone follow-up group, 31 of 54 patients (57%) reported cold intolerance. In the face-to-face group, 21 of 33 patients (61%) reported cold intolerance. The demographics of the two groups are shown in Table 1, and there was no difference between the two groups in any items.

OPEN IN VIEWER

Table 1.

Comparisons of demographics between telephone group and face-to-face group.

Variable	Telephone group (n = 54)	Face-to-face group (n = 33)	p-value
Age (years)	40 (SD 13)	40 (SD 9)	0.836
Range	16–62	23–55
Sex			1.000
Male	41	25
Female	13	8
Follow-up time (months)	35 (SD 12)	33 (SD 15)	0.456
Range	14–61	17–58
Specific digit			0.829
Index	19	14
Middle	20	11
Ring	14	8
Little*	1	0
Type of injury			0.153
Sharp instrument	12	3
Blunt instrument	42	29
Chemical	0	1
Cold intolerance			0.565
Yes	31	21
No	23	12

*These items were excluded from statistical analysis because there were too few cases.

The mean CISS score was 28 marks (range 4–66). The distribution of scores in the patients is shown in Figure 1. Table 2 shows that the triggering temperature grade was correlated with the CISS score, but there was no correlation with other variables, including the sensory recovery in the flaps in the face-to-face group.

OPEN IN VIEWER

Figure 1.

The distribution of CISS scores in the whole study population with cold intolerance. Most patients scored between 11–40 marks.

OPEN IN VIEWER

Table 2.

Correlations between CISS score and other variables.

Variable	Correlation with CISS score	p-value
Age	0.156	0.289
Follow-up time	–0.169	0.643
Triggering temperature grade	0.77	<0.001
SW	0.305	0.190
S-2PD	0.233	0.323

CISS: Cold Intolerance Symptom Severity; S-2PD: static two-point discrimination; SW: Semmes–Weinstein.

Of the 52 patients, four patients partly recovered from the negative symptoms and four patients completely recovered. We found that the mean age of cases with improvement of symptoms (33 years, SD 9) was significantly lower than cases without improvement (43 years, SD 10) (p = 0.008). None complained about worsening of cold intolerance symptoms (Figure 2).

OPEN IN VIEWER

Figure 2.

The frequency of changes in cold intolerance in all patients.

Statistically significant differences were observed between the groups with and without cold intolerance with respect to their ages and the specific digit affected (Table 3). The mean age of the non-cold intolerance group was less than that of the cold intolerance group. The incidence of cold intolerance in the ring finger was statistically significantly lower than in the middle and index fingers (Table 4). In those with cold intolerance, there was no difference between men and women with respect to the CISS scores (p = 0.584).

OPEN IN VIEWER

Table 3.

Comparisons of demographics and follow-up results between cold intolerance group and no cold intolerance group.

Variable	CI group (n = 52)	Non-CI group (n = 35)	p-value
Age (years)	42 (SD 10)	37 (SD 12)	0.048
Range	19–62	16–59
Sex			0.211
Male	37	29
Female	15	6
Follow-up time (months)	34 (SD 15)	33 (SD 14)	0.780
Range	14–61	14–60
Specific digit			0.006
Index	26	8
Middle	19	12
Ring	7	14
Little*	0	1
Dominant hand affected			0.237
Yes	26	13
No	26	22
Type of injury			0.273
Blunt instrument	44	27
Sharp instrument	7	8
Chemical*	1	0
Smoker			0.170
Yes	28	24
No	24	11

CI: cold intolerance.

*

These items were excluded from statistical analysis as there were too few cases.

OPEN IN VIEWER

Table 4.

Comparison of the three different finger groups.

Variable	Index (n = 34)	Middle (n = 31)	Ring (n = 21)	p-value
Age (years)	41 (SD 10)	42 (SD 12)	36 (SD 11)	0.195
Range	19–59	19–62	16–54
Sex				0.280
Male	25	18	16
Female	9	13	5
Follow-up time (months)	34 (SD 15)	31 (SD 13)	42 (SD 12)	0.513
Range	15–61	14–58	14–60
Cold intolerance				PI-R = 0.002
				PM-R = 0.048
				PI-M = 0.185
Yes	26	19	7
No	8	12	14
Donor side	Ulnar	Ulnar	Ulnar
Dominant artery	Ulnar	Ulnar	Radial
Sacrifice of the dominant artery	Yes	Yes	No
Incidence rate	76%	51%	33%

I-R: comparing index finger and ring finger groups; M-R: comparing middle finger and ring finger groups; I-M: comparing index finger and middle finger groups.

Table 5 shows the CISS questionnaire score and number of patients in the groups with different grades of triggering temperature. There were significant differences between the CISS scores in the different grades (p < 0.05). Coldness, numbness and pain were the symptoms of cold intolerance that patients most commonly complained of (Table 6). A 44-year-old man, who had had a left ring finger pulp defect, complained of skin cracking of the flap and the surrounding skin during winter. However, no similar symptom occurred on his other fingers.

OPEN IN VIEWER

Table 5.

CISS questionnaire scores of the groups with different temperature grades.

Grade	Temperature (°C)	Number of patients	Average CISS score (SD, range)
Grade 1	<0	20	17 (8, 4–33)
Grade 2	0–5	11	23 (6, 12–32)
Grade 3	6–10	13	41 (12, 22–66)
Grade 4	>10	4	54 (9, 47–66)
Total		48	28 (16, 4–66)

CISS: Cold Intolerance Symptom Severity.

OPEN IN VIEWER

Table 6.

The first and second complaints from the open-ended questions about cold intolerance symptoms.

Symptom	First	Second	Total (%)
Coldness	30	9	39 (50%)
Numbness	6	15	21 (27%)
Pain	6	4	10 (13%)
Swelling	2	0	2 (3%)
Stiffness	3	1	4 (5%)
Colour change	0	1	1 (1%)
Flap cracking	1	0	1 (1%)

Discussion

We found that 60% of those treated with a reverse digital artery flap transfer for fingertip or pulp defects in the hands suffered from cold intolerance; this is lower than after nerve injury (80%) (Irwin et al., 1997) and digital replantation (almost 100%) (Glickmann and Mackinnon, 1990; Morrison et al, 1978), but higher than after hand factures (38%) (Nijhuis et al., 2010) and vascular injury (41%) (Klocker et al., 2012). Similar to previous reports (Stokvis et al., 2009), the severity of the symptoms mostly remained unchanged. Fortunately, no patients had worsening symptoms over time and about 15% recovered from the symptoms partly or completely; these patients tended to be younger.

The duration of symptoms in patients who develop cold intolerance is controversial. Morrison et al. (1978) reported a slow improvement in symptoms over 10 years after injury. In contrast, a complete recovery after 1 year (Earley and Watson, 1984) and an improvement after 2 or 3 years have been presented (Backman et al., 1993; Craigen et al., 1999; Glickmann and Mackinnon, 1990; Urbaniak et al., 1993). There was a limited follow-up time in the patients we studied, so the natural history in this group is uncertain.

A connection between age and the likelihood of developing symptoms was identified, as the younger patients were less likely to have cold intolerance. However, there was no relationship between age and CISS score (Table 2).

Ruijs et al. (2007) studied patients with nerve injury and found that the CISS scores were significantly higher in women than in men, but we failed to identify any gender influence on the CISS scores.

Irwin et al. (1997) demonstrated that patients were more likely to develop cold intolerance if they smoked and less likely to if they suffered a sharp injury. However, this was not the case in the current study (Table 3).

There were significant differences in cold intolerance in different fingers (Table 4). The prevalence of cold intolerance in the ring finger was lower than in the index and middle fingers. Our previous work found a similar tendency, but in a smaller sample size (Chen et al., 2015). We designed the flap on the index, middle and ring fingers on the ulnar side so that the flap was located in the concealed area. The dominant arteries of the index and middle fingers are the ulnar digital arteries, whereas the dominant digital artery is on the radial side of the ring finger, so the blood supply of flap in ring finger is from radial digital artery which is the dominant artery. See Table 4. The tendency for cold intolerance to occur in the fingers in which the dominant artery was sacrificed suggests that the blood supply is an important predisposing factor for cold intolerance. This view is supported by the work of Gelberman et al. (1978), who have shown that cold intolerance is more likely to occur in an injured digit when the digital pulse pressure is <75% of that in an uninjured finger. Brown et al. (1986) and Tark et al. (1989) have also demonstrated that uncomfortable symptoms are ameliorated in replanted digits by induced vasodilatation or repair of both digital arteries instead of one. Although we prefer a reverse digital artery flap based on the ring finger rather than the other fingers, because of the lower prevalence of cold intolerance, we no longer recommend flap transfer in patients who will encounter a cold environment, because of the likelihood of cold intolerance.

Several limitations should be acknowledged in our study. First, collection of details about the initial injuries and surgical procedures was done retrospectively. Second, only 33 patients accepted a face-to-face interview with measurement of sensory recovery. Given the relatively small number of patients returning for testing of sensory recovery, we were unable to demonstrate any relationship between sensory recovery and cold intolerance. Third, the follow-up time (14–61 months) was too short to allow changes in the pattern of cold intolerance to be determined. Fourth, because of a lack of cases involving the little finger, we could not statistically compare it with other fingers.