British Journal of Plastic Surgery (2005) 58, 1131–1135

Donor site morbidity in cross-finger flaps*

H. Koch*, A. Kielnhofer, M. Hubmer, E. Scharnagl

Division of Plastic Surgery, Department of Surgery, Medical University of Graz, Auenbruggerplatz 29,
A-8036 Graz, Austria

Received 30 September 2004; accepted 28 April 2005

KEYWORDS Summary As relevant literature is scarce, this study was undertaken to assess the
Cross-finger flap; donor site morbidity of cross-finger flaps. It included 23 patients who had undergone
Donor site morbidity; reconstruction of a finger defect with a cross-finger flap. Any additional trauma to
Total range of motion; the donor finger was an exclusion criterion. Split thickness skin grafts were employed
Pinch grip for donor site closure in 13 cases, full thickness skin grafts were used in 10 cases.
Follow-up time averaged 83 months. Active and passive total range of motion of the
donor finger and maximal pinch grip strength in kilopascals were measured. Both
parameters were compared to the corresponding finger of the other hand. The donor
site scar was evaluated for instability and pain in the donor finger was determined
subjectively with a visual analogue scale. Cold intolerance and the cosmetic
appearance of the donor site were also assessed.
Active total range of motion of the donor fingers averaged 1568. Average active
total range of motion of the contralateral control fingers was 173.68. There was a
significant difference between the donor fingers and the control fingers (pZ0.03) but
not between split thickness and full thickness grafted donor sites (pZ0.91). Grip
strength was significantly impaired in the donor fingers (pZ0.03), but there was no
significant difference between split thickness and full thickness grafted donor sites.
Subjective cosmetic evaluation by the patients revealed significantly better results
for full thickness grafted donor sites. Donor finger pain averaged 2.4 with a range of
0–8. Five of the 13 patients with split thickness grafted donor sites and two of the 10
patients with full thickness grafted donor sites mentioned cold intolerance.
In conclusion, the cross-finger flap is a secure and valuable option. There is,
however, significant donor site morbidity. Our results suggest that alternative
solutions should also be considered and if a cross-finger flap is employed, donor sites
should be closed with full thickness grafts.
q 2005 The British Association of Plastic Surgeons. Published by Elsevier Ltd. All
rights reserved.

*
The work was presented at the 34th Meeting of the Vereinigung der Deutschen Plastischen Chirurgen, October 1–4, 2003, Freiburg
(Poster) and the 42nd Meeting of the Austrian Society for Plastic, Aesthetic and Reconstructive Surgery, October 7–9, 2004, Innsbruck.
* Corresponding author. Tel.: C43 316 385 4685; fax: C43 316 385 4690.
E-mail address: horst.koch@meduni-graz.at (H. Koch).

S0007-1226/$ - see front matter q 2005 The British Association of Plastic Surgeons. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.bjps.2005.04.047
1132
Since its introduction in the literature,1 the cross-
finger flap has gained wide acceptance in recon-
structive hand surgery, due to its ease of dissection,
its anatomical security and the provision of soft and
pliable tissue very well suited for reconstruction of
finger defects. Besides these advantages, this flap
also has its disadvantages. There is the apparent
drawback of being a two-stage procedure. As donor
site morbidity is not often addressed with the cross-
finger flap, this study was undertaken to evaluate
the donor site morbidity of this commonly used
procedure.
Materials and methods
Between 1985 and 2001, 48 patients underwent
defect closure on a long finger using a cross-finger
flap. Patients who had sustained any kind of
additional trauma to the donor finger were
excluded from the study. Of the remaining patients,
23 could be examined within the scope of this
study. There were 21 males and two females with
an average age of 30.28 years (1.5–59 years, median
30 years) at the time of surgery. Thirteen defects
were localised on the left and 10 on the right hand.
In seven cases the index finger, in 10 the middle
finger, in four cases the ring finger, and in two cases
the little finger were injured. The defects resulted
from different kinds of trauma (nZ17), thermal
injuries (nZ5) and infection (nZ1). The flaps were
harvested from the second finger (nZ5), from the
third finger (nZ7), from the fourth finger (nZ7),
and from the little finger (nZ4). Split thickness skin
grafts were employed for donor site closure in 13
cases, full thickness skin grafts were used in 10
cases. Follow-up averaged 83 months (24–215
months).
At the time of examination, donor finger pain
was given subjectively by the patient using a visual
analogue scale with 10 grades (0Zno pain, 10Z
maximal imaginable pain). In addition, patients
were requested to subjectively assess the cosmetic
appearance of the donor site. Again, a visual
analogue scale with 10 grades was employed (0Z
no cosmetic impairment, 10Zmaximal cosmetic
impairment). A Mann–Whitney test was used to
evaluate differences between the group of patients
with split-thickness-skin-grafted donor sites (SG)
and the group with full-thickness-skin-grafted
donor sites (FG) concerning pain and cosmetic
appearance. The patients were queried as to cold
intolerance and a raw evaluation of touch sensi-
tivity of the donor site was done; further, the donor
site was evaluated for signs of instability.
H. Koch et al.
Differences between the two groups regarding
cold intolerance, sensibility and instability were
evaluated statistically with Fisher’s exact test.
Physical examination included evaluation of active
and passive total range of motion (TRM) of the
donor finger with a standard hand goniometer, and
of maximal pinch grip strength. This was measured
in kilopascals (kPa) using a Martin vigorimeter
(Gebrüder Martin, Germany). One individual had
to be excluded from strength testing due to a
preceding thumb amputation on the contralateral
side. Both range of motion and maximal grip
strength were compared to the uninjured corre-
sponding finger of the opposite hand; a repeated
measures ANOVA was used for statistical evalu-
ation. A regression analysis was done to evaluate
the influence of the patient’s age at the time of
operation on range of motion and maximal grip
strength of the donor finger.
Results
The results are summarised in Table 1. The overall
mean value for pain on the visual analogue scale
was 2.4 (median 1, range 0–8). There was no
statistically significant difference between the
two groups (pZ0.47). Regarding the subjective
cosmetic estimation of the donor sites, however,
there was a significant difference between the two
groups. The overall mean value for cosmetic
impairment was 3 (median 2, range 0–10). In the
SG the mean was 4.15 with a median of 4 and a
range from 0 to 10. In the FG it was mean 1.6 with a
median of 1 and a range from 0 to 4 (pZ0.04).
Figure 1 Forty-year-old man with cross-finger flap
donor site on the middle finger with flexion deficit and
contour defect.
Donor sites in cross-finger flaps 1133

Table 1 Results
Patient Age TRM D TRM C Pinch Pinch Scar Cosmetic Pain Donor
number (years) active active grip grip instability appearance (VAS) site
(degrees) (degrees) strength strength (VAS) closure
D (kPa) C (kPa)
1 40 145 170 38 46 n 5 0 s
2 30 175 187 10 22 n 5 1 s
3 36 97 155 32 42 y 1 1 s
4 24 170 155 32 36 n 1 0 f
5 24 180 175 34 36 n 1 0 f
6 45 195 180 40 46 n 1 1 s
7 41 143 170 28 28 n 7 7 s
8 37 97 102 30 28 n 10 0 s
9 2 175 175 14 10 n 1 1 f
10 45 150 147 28 26 n 1 1 f
11 43 235 270 10 14 n 2 6 f
12 18 210 230 10 22 n 0 6 s
13 2 191 200 16 16 n 2 0 f
14 47 70 95 12 12 n 3 3 f
15 18 n.e. n.e. 12 50 n 5 7 s
16 41 70 145 12 30 n 0 8 f
17 35 125 185 50 n.e. y 9 1 s
18 17 185 180 32 34 n 4 1 f
19 23 165 190 14 16 n 1 0 f
20 59 120 155 8 16 n 3 8 s
21 21 215 234 30 28 y 4 1 s
22 26 115 120 32 38 n 3 2 s
23 23 205 200 14 16 n 1 0 s
Abbreviations: TRM, total range of motion; D, donor finger; C, control finger; kPa, kilopascals; VAS, visual analogue scale; s, split
thickness skin graft; f, full thickness skin graft.

Figs. 1 and 2 show a patient with a donor site with
significant contour deformity.
Three donor sites showed signs of instability. All
of these were in the SG, but the difference between
the groups was not statistically significant (pZ
0.23). Seven patients reported cold intolerance in
Figure 2 Forty-year-old man with cross-finger flap
donor site on the middle finger with flexion deficit and
contour defect.
the donor finger. Two of these patients had full-
thickness-skin grafted donor sites, five had split-
thickness-skin-grafted donor sites. There was no
statistically significant difference between the
groups (pZ0.4). Two of the patients exhibiting
cold intolerance also described intolerance to heat
in their donor fingers. Impairment of sensibility
distal to the donor site was evident in seven
patients. Five patients out of the SG and two
patients out of the FG were affected. The
difference was not significant (pZ0.4).
Active TRM of the donor fingers averaged 1568
(median 167.5, range 70–235). Average active TRM
of the contralateral control fingers was 173.68
(median 175, range 95–270). There was a significant
difference between the donor fingers and the
control fingers (pZ0.03) but there was no signifi-
cant difference between the SG and the FG (pZ
0.91). Regression analysis showed no significant
influence of the patient’s age on active TRM (pZ
0.09). Figs. 1 and 3 show patients whose donor sites
had impaired range of motion. There also was a
difference between donor and control fingers in
passive TRM, but this was not statistically
significant (donor: mean 173.1, median 180, range
1134
Figure 3 Twelve-year-old boy 10 years after defect
closure on the left middle finger with a cross-finger flap
from the ring finger. Note flexion deficit and colour
mismatch on the donor finger.
76–250; control: mean 185.7, median 183.5, range
100–295; pZ0.07). Interestingly, regression anal-
ysis revealed a statistically significant influence of
age at the time of operation on passive TRM with
more impaired passive TRM at higher ages (pZ
0.02). Maximal pinch grip strength in the donor
fingers averaged 22.2 kPa (median 22, range 8–40),
whereas it averaged 27.8 kPa in the control fingers
(median 28, range 10–50). The difference between
donor sides and contralateral control sides was
statistically significant (pZ0.03). No statistically
significant difference could be found between FG
and SG (pZ0.14). There was no significant influence
of age on pinch grip strength (pZ0.27).
Discussion
The cross-finger flap is a reliable method in soft
tissue reconstruction of the long fingers. The
provision of soft and pliable tissue very well suited
for defects on the fingers, the anatomical security
and the ease of flap harvest make the cross-finger
flap a widely used reconstructive option. Besides
these advantages, however, there is also the
drawback of two surgical steps and a temporary
fixation of donor and receptor fingers that possibly
impairs physiotherapy. Information in the literature
concerning donor site morbidity is somewhat
contradictory.2–4
Paterson et al.4 reviewed 16 patients who had
undergone the reconstruction of finger tip defects
with cross-finger flaps; eight of them had significant
impairment of donor finger joint movement.
Furthermore, Paterson found that patients over
the age of 41 years had a higher degree of donor
finger stiffness than younger patients but he could
H. Koch et al.
not claim statistical significance for that.4 There
are no other data specifically addressing functional
donor site morbidity in the literature. The review
published by Kleinert3 mentions the danger of
contracture if the midlateral line is crossed during
flap harvest. The data found in our study reveal a
significant impairment of total range of motion of
the donor fingers and also a significant decrease in
maximal pinch grip strength of the donor finger to
the thumb. Regression analysis of our results
showed statistically significant influence of patient
age only on passive TRM but not on active TRM and
pinch grip strength. The influence of functional
impairment of the donor finger on global hand
function can only be speculated upon. This is
because a comparison of functional hand scores of
the hand with the injured and the donor finger to
the contralateral hand would be biased by the
functional impairment of the recipient finger.
Furthermore, half of our patients had cosmetic
problems such as contour deformities and colour
mismatch. In Kleinert’s study,3 the danger of
contracture when the midlateral line is crossed
during flap harvest is mentioned and the donor site
grafted with a full thickness graft is designated
acceptable but donor site morbidity was not
explicitly assessed. In a review by Kappel et al.,2
donor site morbidity is only mentioned from an
aesthetic point of view: self estimation by the
patients revealed good or excellent aesthetic
results in two-thirds of the cases.
Paterson4 reported significant donor site morbid-
ity without finding a difference between split and
full thickness grafts and, therefore, advises the use
of split thickness grafts to reduce operating time.
Our results also do not suggest functional differ-
ences between donor sites closed with split
thickness skin grafts and donor sites closed with
full thickness grafts. There are, however, differ-
ences between the two groups concerning cosmetic
appearance as subjectively estimated by the
patients: donor sites covered with full thickness
grafts resulted in statistically significantly less
cosmetic impairment. Furthermore, signs of
instability only occurred in donor sites covered
with split thickness grafts in this study. Thus, in
contrast to Paterson’s publication,4 our results
indicate that the best coverage for donor sites of
cross-finger flaps would be a full thickness skin
graft. This is in accordance with Kappel et al.,2 who
found good cosmetic results with donor site cover-
age by full thickness grafts and, therefore, rec-
ommend this method of coverage. Kleinert et al.3
advocate the use of full thickness grafts as well.
Our results must of course be interpreted with
caution as ours was a retrospective study with a
Donor sites in cross-finger flaps
limited number of patients. There is nonetheless
strong evidence that cross-finger flaps cause
significant donor site morbidity. This morbidity
includes cosmetic impairment as well as cold
intolerance and donor site pain. Furthermore, and
more importantly, this donor site morbidity has a
functional aspect with reduced range of motion as
well as impairment of pinch grip strength of the
donor finger. This is of special importance as the
flap is usually used for a neighbouring finger that is
already injured, i.e. there is a danger of further
impairment of hand function by the cross-finger
flap. The fact that physiotherapy is limited by the
necessity of temporarily fixing the two fingers to
each other adds to this. Altogether, we might
conclude that in addition to the obvious disadvan-
tage of being a two-stage procedure, the cross-
finger flap also holds the drawback of significant
donor site morbidity. Therefore, especially in cases
with severely traumatised neighbouring fingers, we
recommend including other reconstructive options
such as dorsal metacarpal flaps,5 arterialised
venous flaps6 or other small free flaps7,8 in the
therapeutic considerations.
Acknowledgements
The help of Prof P. Rehak, Head of the Division of
1135
Medical Engineering and Computing, Department of
Surgery, Medical University of Graz with statistical
analysis of the data is acknowledged.
References
1. Cronin TD. The cross finger flap. An new method of repair. Am
Surg 1951;17:419–25.
2. Kappel DA, Burech JG. The cross finger flap. An established
reconstructive procedure. Hand Clin 1985;1:677–83.
3. Kleinert HE, McAlister CG, MacDonald CJ, Kutz JE. A critical
evaluation of cross finger flaps. J Trauma 1974;14:756–63.
4. Paterson P, Titley OG, Nancarrow JD. Donor finger morbidity
in cross-finger flaps. Injury 2000;31:215–8.
5. Keramidas E, Rodopoulou S, Metaxotos N, Panagiotou P,
Iconomou T, Ioannovich J. Reverse dorsal digital and
intercommissural flaps used for digital reconstruction. Br
J Plast Surg 2004;57:61–5.
6. Koch H, Scharnagl E, Schwarzl FX, Haas FM, Hubmer M,
Moshammer HE. Clinical application of the retrograde
arterialized venous flap. Microsurgery 2004;24:118–24.
7. Inada Y, Tamai S, Kawanishi K, Omokawa S, Akahane M,
Shimobayashi M, et al. Free dorsoulnar perforator flap
transfers for the reconstruction of severely injured digits.
Plast Reconstr Surg 2004;114:411–20.
8. Pelzer M, Sauerbier M, Germann G, Trankle M. Free ‘kite’ flap:
a new flap for reconstruction of small hand defects.
J Reconstr Microsurg 2004;20:367–72.