Introduction
Fingertip amputations account for two-thirds of all hand injuries in the pediatric
population with door-related crush injuries being the most common mechanism.[1]
[2] Several short-term complications can arise following the injury including cold intolerance,
hypersensitivity, residual pain, and infection. Poor sensory recovery, scar retraction,
and hook nail deformity form part of the long-term sequelae observed in fingertip
amputations, resulting in functional disability and becoming a psychological burden
to carers and patients.[3]
Various surgical techniques such as replantation have been described to maintain normal
function and architecture of the fingertip. While replantation has advantages in re-establishing
immediate blood flow, the distal nature of amputations within the distal phalanx makes
this technique technically challenging in children.[2] Other limitations of replantation include cost, time efficiency, and increased operating
times.[4] Nonmicrosurgical techniques such as healing by secondary intention using various
dressings, reconstruction with local or regional flaps as well as repositioning of
the amputated part as a composite graft have become an alternative and popular method
in recent times.
Paucity in guidelines on the use of composite grafts for fingertip amputation has
left the decision for choosing this technique to remain clinician based.[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12] For example, composite graft technique has been a favored method in most children's
hand trauma units when the amputated part is well preserved and in a condition that
can be used for reconstruction.[5] However, evidence regarding what is defined as “condition for reconstruction” as
well as predictive factors affecting long-term outcomes is sparse.[13] Furthermore, the assessment of outcomes has remained heterogenous in the literature,[13] making it difficult for both clinicians and patients to understand outcomes of composite
grafts as well as difficulty in managing expectations from patients and their families.
This article aims to explore the outcomes of composite grafts in fingertip amputations
in children as well as the contributing factors such as age and level of amputation
that may affect the overall result.
Methods
The study protocol was registered with PROSPERO (CRD42022316590) and was reported
via the Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA)
guidelines. The following databases were used for the literature search from database
inception to May 20, 2023: MEDLINE (PubMed), Embase, Google Scholar and Cochrane Database
of Systematic Reviews. [Table 1] shows the combinations of MeSH search terms used in each database. Keywords such
as “fingertip,” “digital tip” “digit,” “finger,” “thumb,” “amputation,” “injury,”
“replantation,” and “composite grafts” were used. Forward and backward citation searching
was performed for all included and excluded studies as well as relevant systematic
reviews by reviewing the reference lists of each study and identifying articles that
cited said study in Web of Science.
Table 1
MeSH terms used for literature search
|
Concept
|
MeSH term
|
|
Finger
|
‘’anastomosis, surgical,” “graft survival,” “microsurgery,” “surgical flaps,” “surgery,
plastic,” “reconstructive surgical procedures,” “surgical procedures, operative,”
“surgery,” “debridement,” “conservative treatment,” “wound healing,” “therapeutic
irrigation”
|
|
Phalanges of fingers
|
‘’anastomosis, surgical,” “graft survival,” “microsurgery,” “surgical flaps,” “surgery,
plastic,” “reconstructive surgical procedures,” “surgical procedures, operative,”
“surgery,” “debridement,” “conservative treatment,” “wound healing,” “therapeutic
irrigation”
|
|
Thumb
|
“anastomosis, surgical,” “graft survival,” “microsurgery,” “surgical flaps,” “surgery,
plastic,” “reconstructive surgical procedures,” “surgical procedures, operative,”
“surgery,” “therapeutic irrigation,” “wound healing,” “conservative treatment,” “debridement”
|
|
Finger injuries
|
“anastomosis, surgical,” “graft survival,” “microsurgery,” “surgical flaps,” “surgery,
plastic,” “reconstructive surgical procedures,” “surgical procedures, operative,”
“surgery,” “debridement,” “conservative treatment,” “wound healing,” “therapeutic
irrigation”
|
All titles and abstracts retrieved from the databases were downloaded and duplicates
were removed. Full-text articles were retrieved and reviewed independently in the
case of discrepancy concerning inclusion/exclusion of articles into the study. The
exclusion criteria consisted of adult population (over the age of 18), secondary reconstructions
after initial healing, replantations of fingertip, flap reconstruction of fingertip,
pocket flap-grafts, and conference abstracts. The inclusion criteria consisted of
human studies, fingertip (distal phalanx), composite grafts, and English written studies.
Data was extracted using a modified and custom form by four authors. Each author extracted
data from an equal number of included studies. Data was analyzed according to the
Patient, Intervention, Comparison and Outcome (PICO) format. The included studies
were analyzed for patient demographics and outcomes. The patient demographics analyzed
included age, sex, which finger was injured, level of injury (depending on classification
used), type of injury, and number of fingertips. The outcome data extracted was divided
into primary outcomes, secondary outcomes, and complications. Primary outcomes consisted
of graft take, and secondary outcomes consisted of length of follow-up, functional
outcomes, patient satisfaction, and effect on quality of life.
The Risk Of Bias In Non-randomized Studies of Interventions (ROBINS-I) tool was used
to evaluate the risk of bias which is represented visually using robvis tool in [Fig. 1].[14]
Fig. 1 Risk Of Bias In Non-randomized Studies of Interventions (ROBINS-I Tool) assessment
for included studies.
Results
The initial search yielded 776 articles for review after eliminating duplicates. After
full-text screening, 12 articles (six retrospective reviews, two prospective case
reviews, two cohort studies, two case series) published between 1997 and 2020 were
included for full-text analysis. The result of this search and yield of articles is
represented as a PRISMA flowchart in [Fig. 2].
Fig. 2 Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) flowchart
of articles.
Demographical Data
A total of 735 composite grafts were performed with the middle finger (n = 73, 9.9%) followed by ring finger (n = 63, 8.6%) being the most common site of fingertip amputation requiring repair with
this technique. Eight studies specified the age of their pediatric population, and
the mean age of the patients was 4.1 years with the range of 0 to 17 years. From the
seven articles that specified sex of the patients who had composite grafts, 225 males
and 170 females were found. The most common mechanism of injury was crush type injury
(n = 333, 85.2%) followed by laceration (n = 18, 4.6%), from the eight articles that specified this data.
Among the 11 articles that classified the level of injury using a classification,
the following were used: Ishikawa (n = 158), modified Ishikawa (n = 293), Das and Brown (n = 6), Hirase (n = 3), with Imaizumi et al, Son et al, and Heistein and Cook, using their own classification
(n = 17, n = 15, n = 19, respectively).[9]
[15] The various classifications of fingertip injury are shown in [Fig. 3] for comparison.[16]
[17]
[18]
[19]
[Table 2] shows the descriptions for all classifications of fingertip injury found in this
study.
Fig. 3 Anatomical comparison of the various classifications used for fingertip amputations.
DP, distal phalanx.
Table 2
Descriptions for fingertip amputation classifications in this study
|
Classification
|
Description
|
|
Ishikawa/Modified Ishikawa (MI)
|
Subzone I—from tip of fingertip to midpoint of nail
IA (MI)—from tip of the fingertip to the distal end of the nail
IB (MI)—from the distal end to the midpoint of nail
Subzone II—from midpoint of the nail to base of the nail
Subzone III—from nail base to the midpoint between nail base and the DIPJ
Subzone IV—from the midpoint between the nail base and the DIPJ to the DIPJ
|
|
Das
|
Type I—distal pulp (with or without part of nail)
Type II—pulp + terminal phalanx up to distal 1/3
Type III—pulp + terminal phalanx up to distal 3/4
|
|
Hirase
|
Zone DP I—distal to the most distal dividing point of the central digital artery
Zone DP IIA—central digital artery arising from the distal palmar arch of digital artery
Zone DP IIB—distal palmar arch of digital artery
Zone DP III—proximal to the distal palmar arch of digital artery
|
|
Imaizumi
|
Distal type—distal to distal phalangeal tip
Middle type—between Distal and Proximal type
Proximal type—proximal to the lunula
|
|
Son
|
Zone I—distal to lunula
Zone II—at the lunula
Zone III—proximal to the lunula
|
|
Heistein and Cook
|
Zone DP I—area distal to the eponychial fold
Zone DP II—area distal to the DIPJ yet proximal to the eponychial fold
|
Abbreviations: DIPJ, distal interphalangeal joint; DP, distal phalanx; MI, modified
Ishikawa.
Primary Outcomes
Graft Take
Eleven studies reported graft take or survival as an outcome. For the studies where
“complete” graft take was reported as a separate outcome, 17.3% of fingertips overall
achieved this result. For studies that combined “partial” and “complete” graft take
as an outcome, 81.6% of fingertips achieved this result. Eo et al reported graft “success”
in all six of their pediatric patients. This article, however, did not mention graft
take but measured “scab” survival in their pediatric population.[20] This was only described as “good”; therefore, it could not be included in overall
conclusions.
Secondary Outcomes
Patient Satisfaction and Effect of Quality of Life
Three studies reported patient satisfaction. Borrelli et al noted that on average,
patients (n = 100) rated the appearance as “normal” and were satisfied with the cosmetic appearance,
scoring the appearance of the composite graft an average of 3.5/5 with 5 being the
best possible outcome.[5] The authors also reported that most of their patients took 2 to 6 months to return
to normal daily activities following the repair of their fingertips. Urso-Baiarda
et al reported patient satisfaction at 5 years post-surgery, where the patients reported
normal use of the digit and satisfaction with the results of surgery.[21] Uysal et al reported full satisfaction from all three of their pediatric patients
despite shortening of fingers.[22]
Length of Follow-Up
Length of follow-up was reported by eight articles. The mean length of follow-up was
10.1 months across all studies and the range was 0.5 to 96 months. Borrelli et al
(mean follow-up 4.5 months) measured complications and graft survival at follow-up.
Heistein and Cook, (1 week and 12 weeks follow-up) measured clinical progress and
graft survival at follow-up.[9] Eo et al measured color change, presence of infection or hematoma, scab formation
due to insufficient blood supply, and composite graft sensory recovery at follow-up.
Seven of the articles measured only graft take at follow-up.[6]
[11]
[21]
[22]
[23]
[24]
[25]
Functional Outcomes
Two of the included studies mentioned functional outcomes as part of their report.
Borrelli et al reported sensory problems in 16 to 30% of patients, of which most were
due to a tender fingertip or scarring.[5] Butler et al mentioned that their study demonstrated “excellent” long-term functional
outcomes; however, they did not discuss any specific criteria for this conclusion.[6]
Complications
The overall infection rate post-composite graft was 3.8% (n = 28). Interestingly, Butler et al also reported that patients were significantly
more likely to have a postoperative infection if they had an amputation at modified-Ishikawa
Level II than if the amputation was in Level Ia or Ib (22 vs. 6%, p = 0.03).[6]
Nail Abnormalities
Overall, 3.4% (n = 25) of composite grafts presented with a complication of hook nail deformity found
across four studies. The highest rate of hook nail deformity was found in the study
by Butler et al, who reported nail growth abnormalities in 48% of cases, accounting
for 20 patients.[6] This study hypothesized that finger shortening and nail curving cases may be related
to bone nibbling when the bone was exposed.[5]
Fingertip Shortening
Two studies reported fingertip shortening. Borrelli et al reported an average fingertip
shortening between 3.93 ± 2.84 mm in 57% of cases, accounting for 29 patients. The
patient reported outcome of fingertip shortening in this study ranged from 1 to 10mm.[5] Uysal et al reported a mean finger shortening of 6.8 mm.[22]
Factors Affecting Outcomes of Composite Grafts
Level of Amputation
No significant correlation could be identified between graft take and level of amputation;
however, in all studies that used the modified Ishikawa classification and included
data for graft take, a difference between graft take for Levels Ia and III was observed
with amputations at Level Ia showing higher levels of graft success. Overall, “complete”
graft take was observed in 52.6% of fingertips in Level Ia, 50% for Level Ib, 50.9%
for Level II, and 44.4% for Level III. Similar findings were seen in studies using
Ishikawa classification. [Table 3] shows the results of the studies.
Table 3
Study designs with demographical data and associated outcomes
|
Author, year (country)
|
Study design
|
No. of patients (sex)
|
No. of composite grafts
|
Age
|
Classification used
|
Level of injury (n)
|
Mechanism of injury
|
Digit (n)
|
Length of follow-up
|
Graft take
|
Complications
|
|
Borrelli et al 2019 (UK)[5]
|
Retrospective review
|
Male, n = 57 (57.0%)
Female, n = 43 (43.0%)
|
100
|
Range 0.08–15.8 years; mean 4.41 ± 3.98 years
|
Modified Ishikawa
|
Ia, n = 3(3.0%);
Ib, n = 26 (26.0%);
II, n = 42 (42.0%);
III, n = 16 (16.0%)
|
Crush, n = 75 (75.0%); avulsion, n = 13 (14.0%); laceration, n = 12 (12.0%)
|
Little finger, n = 31 (31.0%); ring finger, n = 24 (24.0%); middle finger, n = 21 (21.0%); index finger, n = 18 (18.0%); thumb, n = 6 (6.0%)
|
Range 0.5–96 months; mean 4.65 ± 10.85 months
|
Complete, n = 13 (13.0%); partial, n = 46 (46.0%); failed, n = 41 (41.0%)
|
Infection, n = 17 (17.0%); hook nail deformity, n = 1 (1.0%); PTSD, n = 1 (1.0%); anxiety, n = 1 (1.0%)
|
|
Butler et al 2015 (UK)[6]
|
Retrospective review
|
Male, n = 55 (56.7%)
Female, n = 42 (43.3%)
|
97
|
Range 1–15 years; mean 4.3 years
|
Modified
Ishikawa
|
Ia, n = 12 (12.4%); Ib, n = 51 (52.6%); II, n = 32 (33.0%); III, n = 2 (2.1%)
|
Crush, n = 88 (90.7%); other forms of crush type, n = 6 (6.2%); sharp amputations, n = 3 (3.1%)
|
Did not specify
|
Mean 1.8 months
|
Complete, n = 10 (13.4%); partial, n = 33 (34%); failed, n = 54 (56%)
|
Hook nail deformity, n = 20 (48.0%); infection, n = 11 (11.0%), 1 with complete graft survival
|
|
Eberlin et al 2014 (USA)[23]
|
Retrospective review
|
Male, n = 24 (61.5%)
Female, n = 15 (38.2%)
|
39
|
Range 1–18 years; mean 5.9 years
|
Did not specify
|
Did not specify
|
Crush: closure of a door, n = 24 (51.5%), mechanical device, n = 6 (15.4%), other crush injury, n = 5 (12.8%); sharp laceration, n = 2 (5.1%), sports, n = 1 (2.6%), strangulation injuries, n = 1 (2.6%)
|
Middle finger, n = 15 (38.5%); ring finger, n = 9, (23.1%); little finger, n = 8 (20.5%); index finger n = 6 (15.4%); thumb, n = 1 (2.6%)
|
Mean 4.5 months
|
Complete, n = 3 (7.7%); partial, n = 23 (59.0%); failed, n = 13 (33.3%)
|
Revision procedures, n = 4 (10.0%), all no initial graft uptake
|
|
Eo et al 2009 (Korea)[20]
|
Cohort study
|
Male, n = 1 (16.7%%)
Female, n = 5 (83.3%)
|
6
|
Range 1–5 years; mean 2.5 years
|
Das
|
I, n = 5 (83.3%); II, n = 1 (16.7%)
|
Crush type, n = 5 (83.3%); clean-cut amputation, n = 1 (16.7%%)
|
Middle finger, n = 2 (33.3%); ring finger, n = 1 (16.7%); small finger, n = 2 (33.3%); index finger, n = 1 (16.7%)
|
Did not specify
|
Good with no scab formation, n = 3 (50%); good with scab formation, n = 3 (50%), fail, n = 0 (0%)
|
Did not specify
|
|
Heistein and Cook 2003 (USA)[9]
|
Prospective case review
|
Did not specify
|
19
|
Range 1–17 years
|
Own classification
|
|
|
Did not specify
|
Follow-up at week 1 and week 12 post-injury
|
Graft survival 77.3%
|
None
|
|
Imaizumi et al 2013 (Japan)[24]
|
Case review
|
N/A
|
17
|
Range 20–81 months; mean 3 years and 8 months
|
Own classification
|
Distal, n = 3 (17.6%); middle, n = 13 (76.5%); proximal, n = 1 (5.9%)
|
Crush (did not specify), avulsion (did not specify)
|
Middle finger, n = 8 (47.1%); ring finger, n = 3 (17.6%); little finger, n = 2 (11.8%); index finger, n = 2 (11.8%); thumb, n = 2 (11.8%)
|
Range 11–375 days; mean 80 days
|
11/17 attempted: complete, n = 5 (47.1%), failed, n = 6 (52.9%)
|
Postoperative arterial occlusion, n = 1 (5.88%)
|
|
Karakas and Yuce 2020 (Turkey)[27]
|
Case series
|
N/A
|
185
|
N/A
|
Ishikawa
|
Did not specify
|
Did not specify
|
Did not specify
|
Did not specify
|
Complete and Partial, n = 151 (81.1%); failed n = 35 (18.9%)
|
Did not specify
|
|
Moiemen and Elliot 1997 (UK)[11]
|
Prospective cohort study
|
Male, n = 38 (76.0%)
Female, n = 12 (24.0%)
|
50
|
Range 1–14 years; mean 5.7 years
|
Modified Ishikawa
|
Ia, n = 4 (8.0%); Ib, n = 17 (34.0%); II, n = 21 (42.0%); III, n = 8 (16.0%)
|
Crush, n = 38 (76.0%); other crush injury, n = 9 (18.0%); sharp amputation, n = 3 (6.0%)
|
Did not specify
|
Mean 14.8 months
|
Complete, n = 11 (22.0%); partial, n = 26 (52.0%); failed, n = 13 (26.0%)
|
None
|
|
Murphy et al 2017 (Australia)[25]
|
Retrospective review
|
Male, n = 48 (48.0%)
Female, n = 52 (52.0%)
|
96
|
Range 0–16 years; median age 2.4 years
|
Modified
Ishikawa
|
Ia, n = 16 (16.7%); Ib, n = 36 (37.5%); II, n = 13 (13.5%); III, n = 3 (3.1%)
|
Crush, n = 89 (92.7%); sharp laceration, n = 4 (4.2%); mechanism not recorded, n = 3 (3.1%)
|
Middle finger, n = 26 (27.1%); ring finger, n = 25 (26.0%); little finger, n = 18 (18.8%); index finger, n = 25 (26.0%); thumb, n = 2 (2.1%)
|
Median 4 follow- up appointments; mean 68 days
|
Complete, n = 15 (16.0%); partial, n = 50 (52.0%); failed, n = 31 (32.0%)
|
Hook nail deformity, n = 3 (3%) (all partial graft uptake)
|
|
Son et al 2005 (South Korea)[15]
|
Retrospective study
|
Did not specify
|
15
|
Range 1–15 years
|
Own classification
|
Did not specify
|
Did not specify
|
Did not specify
|
Did not specify
|
Complete, n = 12 (80%); failed, n = 3 (20%)
|
Did not specify
|
|
Urso-Baiard et al 2009 (UK)[21]
|
Retrospective case studies
|
Did not specify
|
108
|
Median 5.9 years; IQR 2.8–13.1 years
|
Ishikawa
|
I (Did not specify)
II (Did not specify)
III (Did not specify)
|
Crush (Did not specify)
Avulsion (Did not specify)
Laceration (Did not specify)
|
Did not specify
|
Mean 5 years
|
Complete or partial in children 88.5%
|
Did not specify
|
|
Uysal et al 2006 (Turkey)[22]
|
Prospective case review
|
Male, n = 2 (66.7%)
Female, n = 1 (33.3%)
|
3
|
Range 1.5–6 years; mean 3.83 years
|
Hirase
|
IIa, n = 2 (66.7%); III, n = 1 (33.3%)
|
Crush n = 3
|
Middle finger, n = 1 (33.3%); ring finger, n = 1 (33.3%); index finger, n = 1 (33.3%)
|
Mean 14 months
|
Complete, n = 3 (100%)
|
Nail deformity in one patient
Finger shortening of 4mm on average
|
Abbreviations: IQR, interquartile range; PTSD, posttraumatic stress disorder.
Age
Age was identified as a factor impacting the results of composite graft in two papers.
Borrelli et al and Butler et al found that children under 4 years old had higher rates
of composite graft survival compared with older children.[5]
[6] Son et al also found higher rates of graft survival in patients who were 6 years
of age and younger compared with patients who were 16 years of age and older, but
this difference was not statistically significant.[15] Butler et al specifically noted that those under the age of four had a significantly
greater likelihood of composite graft survival compared with patients above the age
of four (14 vs. 3%, p = 0.02).[6] On the other hand, Heistein and Cook, Murphy et al, Urso-Baiarda et al, and Eberlin
et al reported no significant correlation between age and composite graft uptake.[9]
[21]
[23]
[25] Moiemen and Elliot reported that there is no evidence behind recommending composite
graft solely for children.[11] Karakas and Yuce offered age-based treatment methods based on the pre-existing literature
findings: composite graft for younger patients and V-Y advancement flap for older
children and adolescents.[26] One possible explanation for this was that younger children were more likely to
experience crush injury.[21]
Mechanism of Injury
Only four studies reported the mechanism of injuries.[6]
[11]
[23]
[25] For crush injuries, a majority were due to fingertips amputated by being caught
in a door (n = 239, 72%) with only six patients with crush injuries due to mechanical devices.
No clear mechanism was provided for sharp lacerations and no mention of mechanism
was provided for avulsion injuries. While Eberlin et al found the mechanism of injury
insignificant, Borelli et al found that crush injuries were significantly more likely
to survive than avulsion injuries in multivariable analysis (odds ratio: 5.430 p = 0.018).
Time since Injury to Composite Grafting
Six studies reported the time from injury to composite grafting. However, only three
studies involving 152 fingertips provided comparable data on survival of composite
grafts in relation with time since injury to intervention,[6]
[11]
[20] where there was a trend for higher survival of grafts when composite grafting was
performed earlier. A mean time of 4.3 hours was associated with complete graft take,
7.1 hours with partial graft take, and more than 9 hours for failed graft take. Of
note however, in four studies,[5]
[6]
[23]
[25] the authors reported no significant difference between time from injury to composite
grafting with graft survival. Moiemen and Elliot's study was the only study to report
statistical significance between grafts surviving when performed less than 5 hours
and when performed more than 5 hours.[11]
Other Potential Factors
In four studies involving 278 fingertips (38%), the fingertips were cooled before
composite grafting was performed. Only two studies mentioned the exact method whereby
the amputated part was placed in saline soaked swabs that was then placed in a sealed
bag in ice and water.[6]
[11] From the comparable data of 146 fingertips from two studies,[6]
[11] 14% of fingertips that were cooled beforehand achieved complete graft take, 40%
achieved partial, and 46% were failed graft take. Of note, Borelli et al's study was
the only study to report a statistically significant result of a higher rate of composite
graft survival when cooled compared with fingertips that were not cooled.[5] No association could be found between presence of fractures (whether clean cut,
splintered or comminuted) and graft take outcomes due to lack of sufficient data and
no bones were shortened for all the composite grafts performed.
Discussion
Composite grafts have become a more common method of fingertip amputation management
in children in recent times.[2] Despite its common use, data with regard to effectiveness of composite grafts as
a treatment option is sparse. There is also paucity in the literature on the determining
factors that may influence outcomes of composite grafts in this population that prevents
evidence-based decisions to be made when patients present with fingertip amputations.
Based on the only consistent outcome that was used across the studies (graft take),
composite grafting was found to be a relatively effective method of managing fingertip
amputations in the pediatric population with a high percentage of grafts presenting
with “partial” and/or “complete” take. Composite grafts were observed to be more commonly
performed in male children less than 5 years old with middle fingertip amputations
due to crush type injury. The authors observed a long mean follow-up time of 10.1
months, but this was due to one of the studies reporting a follow-up time of 96 months.
The authors also observed a relatively low rate of infection post-composite graft
and high patient satisfaction.
From the data available, the authors observed that more distal amputations lead to
better outcomes and reduced complications overall in studies that selected both Ishikawa
and modified Ishikawa as their classification of fingertip amputation, particularly
in the region of Ib/II where the central digital artery in the distal phalanx divides.
This result may be due to the increased vascularity in this area with the terminal
segmental branch and fibrous hiatus branch in close vicinity. Only two articles were
able to find statistically significant correlations where a lower age of patients
(<4 years) leads to better outcomes.[5]
[6] Borrelli et al also found a correlation between injury mechanism and graft take,
where composite grafts performed after crush injuries were more likely to survive.[5] This is concordant with the results observed in our study. We observed an association
of higher composite graft survival with a lower time to operation, but this was based
on small number of studies and fingertips.
A significant limitation observed in this study was the variability in assessing and
reporting graft take due to the lack of standardized assessment for graft success,
particularly in the assessment of “partial” graft take. Some authors described graft
take in three categories: “complete” take, “partial” take, and graft failure, whereas
other authors described it in only two categories: “complete” take and graft failure.
Where reporting was done in three categories, percentage results were sometimes calculated
by combining “complete” and “partial” take. Due to the difficulty in separating the
data consistently, analysis of the results therefore remained difficult. This variability
can be attributed to the differences between the authors' understanding and definitions
of what constituted as “complete” and “partial” graft take. Of note, Butler et al
provided the clearest definition with “complete” graft take as having no areas of
necrotic tissue and “partial” graft take constituting of any graft where there were
patches of necrotic tissue.[6] The choice of classification system also introduced heterogeneity in how the injury
was diagnosed and subsequently treated, leading to a lack of standardized outcomes
as the proportion of successful outcomes could depend on the classification chosen
and the level of fingertip amputation the chosen classification selects. The length
of follow-up was also not constant. For example, assessment of “partial” graft take
could have varied with the length of follow-up with some complications such as scar
maturation and bone development presenting at later periods of time. Imaizumi et al
also reported that the reason for various follow-up periods in their own study was
that some children did not come back to clinic after seeing their fingertips survive,
while others came back for their subsequent fingertip deformity, making the results
difficult to compare and analyze.[24] Finally, the outcomes between young and very young patients may have been different
due to the different tissue demand of oxygen and nutrients. However, the outcomes
of composite grafts could not be differentiated based on age due to the heterogeneity
in how age was reported in the individual studies.
In the study conducted by Jerome and Malshikare, the authors developed a fingertip
injuries outcome assessment score that consisted of measuring nail aesthetics, finger
length, pulp pad, bone consolidation, cosmesis, sensation with two-point description,
pain, range of motion, grip strength, and return to work based on their experience
in managing fingertip amputations.[27] Although the study population was mainly adults who did not undergo composite graft
procedures, this study can serve as a benchmark for the development of a composite
graft outcome assessment tool especially as this scoring system achieved an acceptable
level of reliability and internal consistency on statistical analysis. Forming a standardized
assessment tool for pediatric composite graft outcomes may help the clinician make
a more informed decision, for example, at follow-up with regard to deciding on whether
further intervention is required.
Conclusion
Composite grafts for fingertip injury in pediatric patients can be considered as a
relatively effective method of treatment with low rates of complications such as infection
and hook nail deformity and high patient satisfaction based on evidence found in this
study. Clinicians should be aware of the increased risk of complications as well as
poorer outcomes associated with more proximal fingertip amputations (notably beyond
Level II on the modified Ishikawa classification) and thus amputations at this level
may warrant other surgical interventions such as microsurgery. Furthermore, based
on the literature, crush type injuries (mostly due to fingertips trapped in doors)
as well as the use of composite grafts in children less than 5 years old were associated
with better outcomes. However, this result is based on heterogeneous existing data
on the use of composite grafting in the pediatric population and thus, well-designed
prospective studies with standardized methods of assessment are required.
